Genetic, but Not Behavioral, Evidence Supports the Distinctiveness of the Mealy Amazon Parrot in the Brazilian Atlantic Forest
Round 1
Reviewer 1 Report
The paper focuses on a complete molecular assessment of the taxonomic arrangement within a formerly monotypic and widespread American Parrot that has recently been split into two sibling species. The authors question the possibility of further splitting a lineage not studied yet from the isolated Brazilian Atlantic Rainforest and using as tools genetic differentiation and differentiation in open-ended learnt vocalizations and their relationship. The paper is overall well presented, adequately solved and the conclusions scientifically sounded. Still, before acceptance, there are several points that the authors should address
-First of all, it seems there is certain confusion between genetic diversity and genetic differentiation (measured as distances). Secondly, it is not clear how the distances are obtained
INTROD
L 115. Why this second goal is general and not restricted to the case of study?
L 119 Change differentiation for distance
L 138 Colombia for Columbia
MATMET
L 140 How many samples are unique to the study?
L 156-161 Sampling and markers details are not part of the Phylogenetic analyses, they should go apart in their section or moved to the former sections. In the sampling, the description should summarize also the samples borrowed from other papers.
L 181 Which evolution model is using the ML reconstruction?
L 191 2.3 'Genetic Diversity? I'm afraid here the concept of genetic diversity and genetic distance are getting confused. The authors are referring to (and measuring) genetic differentiation (=distance) and not to diversity.
L 194 Are you applying the Kimura-2P (K2P) adjustment to which alignment? Why K2P? This is a nice parameter correction for CytB fragments and commonly used to compare distances between similar taxa, but this model should be justified when is used in a combined sequence of multiple fragments.
I suggest estimating genetic distances with K2P only for Cytb sequences and/or estimating the distance again with the combined alignment but selecting with Modeltest the best fitting model to your sequences and applying this (or the most similar one) to estimate your distance matrix.
L 208 Units of the length window?
RESULTS
L 229-258 and further, write all Latin names in italics.
L 240 How can you say this from the topology and without the genetic distance values between lineages? Please rephrase.
L 248 Leave out 'reduced' (if this is reduced, how will be the one based on cytb alone?)
L 253 network lacks instead of 'networklacks'.
L 253 Why is the 'virenticeps' lineage missing in the network'? please explain,
L 272 Again, change divergence/distance for 'diversity' and see comments in MatMet about this calculation
L 277 The haplotypes description should go with the Networks description.
L 286 PCA: The percentage of the variance explained by PC1 and PC2 is quite low (barely 50%) What's about the other components explanation? Would make any sense to plot other combinations like PC1 vs PC3 or PC2 vs PC3?
DISCUSSION:
L 313 It is confusing taxonomically speaking to talk about subspecies without referring to any particular species. I'd suggest instead referring to 'lineages' when talking about a differentiated and named taxon linked to a particular region.
L 332 ...' perhaps subspecies' this is confusing. If you think it deserves a subspecific recognition, you should say .. distinct population that warrants its taxonomic description as subspecies within A. farinosa.
L 333-352 Genetic distances are only comparable when they refer to the same molecular fragment and using the same (or similar) evolutionary model. You need to explain which fragments and models are behind the values for the distances between taxa in all these papers.
L 368-398 The usage of a character that has not evolutionary/genetic underlying relation with the genetic makeup of the species, like an open-ended learning vocalization is expected to have a poorly predictable taxonomic value. Still, as the authors show with the cited references this character works for some other birds. I suggest, therefore, to keep the hypothesis at a lower expectation level and present this hypothesis for this particular study case and avoid generalization as an inspection of a valid general methodology to investigate 'potential cryptic species' as is presented (also in the introduction).
L 400 Wouldn't help the protection of the unique Atlantic lineage of the species its proper taxonomic description and official recognition as a valid endemic subspecies??. I leave this reflection
Table S1:
Headline: Use either 'gene' or 'region' but not both (I'd prefer the first one). What is the b) note?
Fig.2 a)
-The name of the lineage 'virenticeps' is missing in the tree.
-The selection of colours is a bit unfortunate and the haplotypes of Atlantic lineage and the inormata lineage are hard to tell apart.
-Why the lineage 'kawaii' is showing the same formatting code as the ingroup, despite being an outgroup? Please change since it's confusing.
Fig. 4
Make the side colour scales smaller. They give little extra information.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Review of Diversity-1236486
Genetic, but not behavioral, evidence supports the distinctiveness of the mealy amazon in the Brazilian Atlantic Forest
This manuscript describes a study investigating the potential genetic and behavioural distinctiveness of a population of mealy amazon parrots in the Brazilian Atlantic forest. As part of a cryptic species complex, this work is important for conservation and management. While the authors found some genetic differentiation between this population and others closest in geographic proximity, it was not enough to delineate the population as a separate species. In addition, vocal learning likely contributes to the high variability in calls within and between populations, but would also not support species status. Overall I found the manuscript well-written and robust. The organization made a lot of sense to me, and I was really impressed with it. Minor comments are below.
Title and/or keywords: I would suggest putting “parrot” in the title, for anyone not familiar with the species (as it really is not obvious that it is a bird, especially for non-ornithologists who may still be interested in the subject matter). I also think it is a *tad* misleading, as there is limited distinctiveness (i.e. not species-level).
Lines 24-27 (Abstract) – This sentence is long and somewhat convoluted. Could you simplify? I am having a difficult time actually understanding what was found with respect to calls.
Line 55 – Should probably be “correlate with genetic divergence”
Line 79 – Define “large bodied” (can just be mass range in brackets, to give context).
Line 161 – Looks like a space is missing between “then PCR”
Line 238 and elsewhere – I just notice that here, you don’t italicize the short forms for the species and subspecies names, but you do earlier in the manuscript. It would be good to be consistent.
Line 253 – Looks like a space is missing between “work lacks”
Figure 2 – Could you put all the “a, b, c” labels inside the frames? Right now it looks messy. They don’t necessarily have to all be in the same position, but I think at least having them lined up nicely and all either in or out of the frames would look a lot more professional.
Line 372 – Should be “songbird species”
Line 374 – There are also open-ended learners among songbirds as well, just make sure you don’t sound like you are delineating the groups based on this trait with your wording.
Line 376 – It’s not completely clear why this would be the case – could you elaborate a bit on why open-ended learning would result in acoustic traits not correlated with genetic traits? You still have things like morphological features that are genetically influenced that can constrain vocalizations, and you still have learning from each other, which if you are in an insular population would still result in a population-wide ‘dialect’ that could be distinct from other geographical populations. The reasoning I’m guessing might occur if there is selection to vary one’s vocalizations throughout their lifetime to be more different from each other, meaning there is selection for within-population call variability. Under morphological and physiological constraints, there would only be so much variability across a whole species, such that if each population undergoes selection for individual distinctiveness you would not get inter-population divergence or specificity of calls, but rather high intra-population variability in every population that would likely start to cover the same acoustic space. It would definitely be worth going into this in more detail to really convey the reasoning behind the theory you have here (or even if you have other ideas!).
Line 386 – I assume you mean after they have dispersed from their natal area? Otherwise this doesn’t really make sense.
Line 393 – How would citizen science inconsistency actually have influenced these results? Is there a mechanism here? Seems like a bit of an unjustified blanket statement. If this is the case (you give a kind-of reason in line 396) would there be a way to fix this? Quality filters maybe? If so, why didn’t you use this?
Line 398 – I would argue there are other facets of this that would likely play a part, e.g. are the open-ended learners mimics of local fauna? Or are they leaning from each other? What is selected for? Do they change their vocalizations constantly or do they try and keep vocalizations stable during, say, a season so they are individually recognizable? How much is known about the function of the vocalizations of this species versus others, and could any of these characteristics or uses or patterns be important in this context?
Author Response
Please see the attachment.
Author Response File: Author Response.docx
