The paper titled “Multiple evolutionary transitions of reproductive strategies in a phylum of aquatic colonial invertebrates” by Heather et al. provides valuable insights into the evolution of parental care and incubation strategies among bryozoans, particularly within the Cheilostomatida order. Thy phylogenetic analysis based on mitogenome data and nuclear ribosomal genes (18S and 28S) includes many interesting newly sampled cheilostome bryozoans, and the analysis is well-structured. In addition, the authors have effectively conducted ancestral character estimations. Generally, the conclusions drawn are well-supported by the presented results. I therefore only have a few comments/suggestions edits to try to make the paper even clearer.
Line 84 “Bryozoa is a medium-sized phylum with 6,716 known extant species” Please add approximately before the number of species.
Line 84-85 “of aquatic, colonial, filter-feeding lophotrochozoans” the authors say that all bryozoans are colonial, but some taxa are pseudo-colonial or solitary including aethzoids and monobryozoids.

No justifications have been given for taxon selection, especially for those assembled from the publicly available data. Additionally, the authors mention that both partial and complete mitogenomes were generated for the newly sampled taxa, but the authors did not provide an explanation for why some samples were sequenced partially while others were sequenced completely.

It is important to indicate in the methods section of the main text how many samples have been sequenced in this study (i.e. the number of samples with complete mitogenomes and the number of samples with partial mitogenomes) Additionally, it make sense to mention the total number of taxa that were included in the phylogenetic analysis in methods section also not only in the results.

Lines: 199-200, “Alignments were carried out in MAFFT”
Please include the setting of MAFT in the methods of the main text. This would enhance the clarity and reproducibility of the study's methods.

Lines: 202-203, “The inclusion set provided by Gblocks was reversed into an exclusion set using Mesquite v. 3.51 [41]” this sentence is not very clear.

Since the phylogenetic analyses include many taxa with partial sequences, it would be worth assessing the effect of the missing data on the phylogenetic relationships.

Line: 227, the term ‘introvert’ is long outdated in bryozoans and should be replaced by tentacle sheath. also, this kind of brooding is not present in all Ctenostomata.

Line 255-256: The authors indicate that “Posterior predictive simulations revealed significant compositional heterogeneity in the mt PCGs and 12S+16S rRNA partitions (X2, p < 0.0)” and the NDCH-C2 model was used to increase the fit of the model to the data, however, there was an issue with convergence. Have the authors attempted to run the analysis with other heterogeneous models, such as CAT models (CAT-GTR), as implemented in PhyloBayes? CAT models have been demonstrated to better describe and fit heterogeneity across the phylogeny compared to other heterogeneous models for specific datasets (Whelan & Halanych, 2017). It would be beneficial if the authors considered employing CAT models, particularly (CAT-GTR) in their analysis.

It is important to indicate in the main text of the manuscript whether the different phylogenetic reconstruction methods used in this study have resulted in highly consistent tree topologies or if there are significant differences in the phylogenetic relationships among the different methods. It would be beneficial to present the results obtained from all the employed methods (if the topologies are not consistent across methods), not only relying on the "MrBayes5D topology (figure 2)" for reporting the findings.

One cannot distinguish between the newly generated sequences and those that were taken from publicly available data in Figure 2. It would make the figure clearer if this added. Additionally, localities are shown for some taxa without explanation in the figure legend (is it only shown for the newly generated data?). Moreover, it is not clear why the “MEMBRANIPORIDAE”, “SINOFLUSTRIDAE” and “ELECTRIDAE” are shown in blue while the remaining taxa are shown in black. Clarifications and appropriate labeling of the data in the figure would improve its overall clarity and interpretability.