Blind measurements did not confirm effects of forest fragmentation on fluctuating asymmetry of a tropical butterfly Morpho helenor

Re-evaluation of photographs of the tropical butterfly Morpho helenor from a previous study (Pignataro et al. 2023) revealed that its conclusion regarding increased wing fluctuating asymmetry in forest edge habitats compared to forest interior habitats could not be replicated. This discrepancy likely arises from (i) original measurements not being conducted blindly, (ii) insufficient photograph quality hindering accurate landmark selection, and (iii) a lack of detailed description of the measurement protocol. The likelihood of false positive discoveries within the published data concerning the impacts of environmental stress on the fluctuating asymmetry of plants and animals is probably higher than previously assumed. Supplementary Information The online version contains supplementary material available at 10.1007/s00114-024-01913-9.

wings across all 60 specimens of Morpho helenor studied by Pignataro et al. (2023) and is therefore unlikely to affect FA values calculated from these measurements.
At the same time, I surprisingly discovered that many of the data from Pignataro et al. (2023) are not actual measurements but estimates (approximations).For example, the length of right and left wings reported by Pignataro et al. (2023, and pers. comm.) for the specimen coded 'edge 2' exceeds the maximum distance between wing base and apex (as shown on an image) by 0.66 and 1.44 mm.To obtain these large values, the measurer should have selected the basal landmark outside the wing image.This action is neither reported nor illustrated by Pignataro et al. (2023): figure 1 in their work shows that the basal landmark in the right wing is placed at the broken wing margin rather than outside the wing image.
The placement of landmarks outside the wing image applied by Pignataro et al. ( 2023) is particularly uncertain in wings with damaged apical and/or basal parts (Fig. S1).For example, the length of the left wing of the specimen coded 'interior 10' is reported by Pignataro et al. (2023, and pers. comm.) as 58.87 mm, whereas the maximum wing size that could be measured from this image (Fig. S1c) is 54.85 mm.The difference between these values (4.02 mm) is eight-fold greater than the mean absolute value of the difference in the length of the right and left wings reported by Pignataro et al. (2023).Needless to say, the placement of a landmark outside the measured structure is subjective and therefore likely to enhance the impact of confirmation bias on the conclusions of a study.
To avoid the need to use approximations instead of actual measurements, I defined forewing length as the maximum distance between the base of the Sc stem and the external wing margin at the apex, whereas wing width was measured as the distance between the base and apex of the A1+A2 stem.The bases of these stems were defined as the points of their articulation with the 2nd and 3rd axillary sclerites, respectively.The positions of these landmarks are close to the landmarks used by Pignataro et al. (2023), and I therefore believe that both sets of landmarks are equally suitable to quantify wing FA.
The bases of both Sc and A1+A2 stems in many of the analyzed images could be identified due to an abrupt change of colour between the shining vein and the matte background.If this change was not clearly visible, then I positioned the basal landmarks at the level of wing articulation to thorax, as identified from the shape of wing margins, at about 1/3 and 2/3 distance between wing margins for Sc and A1+A2 stems, respectively.The distal landmark for wing length measurements was selected near the wing apex by moving the measurement point around the external wing margin and recording the maximum value of the distance from the basal landmark.The distal landmark for wing width measurements was selected at the point of A1+A2 arrival at the external wing margin.If this point was not clearly visible, I identified this landmark by projecting visible parts of the A1+A2 vein to the external wing margin.Finally, I excluded from my analysis the specimens (Fig. S1) that, due to wing damage, lacked at least one of the landmarks identified above.
Missing value: one of the two landmarks is missing due to absence of a part of either wing apex or wing base (Fig. 1S).

Table S1 .
Basic statistics on repeated blind measurements of the left and right forewings of