I agree with the comments of the two reviewers and in general also with the response of the authors. I would recommend major revisions of the manuscript before it might be considered for publication.
However, I have some further comments / suggestions for the revision of the manuscript:
Reviewer 2 indicated that the term “andic properties” has not been used properly. I agree with the reviewer. However, the soil was classified as an Andosol requiring the identification of andic properties (or vitric properties). According to WRB there are 3 diagnostic criteria for andic properties (an Alox + ½Feox value of ≥ 2%; and a bulk density of ≤ 0.9 kg dm-3; and a phosphate retention of ≥ 85%.). The authors should comment on that. Furthermore, they have to develop a convincing strategy for using proper terms. The replacement of andic properties by andic soil materials is not the best option because the latter is not really defined. The alternative provided in the response letter “short-range-order constituents” might be used consistently and throughout the manuscript. However, please replace “constituents” by “minerals”.
The whole manuscript is by far too long. Particularly the discussion is sometimes very broad and not all the time focused on the main hypotheses of the paper. One reason for that might be the combined “Results and Discussion” section. The authors should think about the advantages to change the whole structure of the manuscript (separated results and discussion sections).
I give you some examples in the manuscript where the need for reduction is very obvious to me: lines 277-285, 328-339, 353-378, 401-415, 436-467, second part of the section 468-490
The acronyms for the two mineral clusters are not very straightforward. Using ALL for allophanic but also for all samples is misleading and a potential source of confusion. Do you really need an acronym? I would suggest to use the full names or more appropriate acronyms. There is no need to use these acronyms in tables – there is sufficient space for the full name.
More specific comments:
In the Introduction (line 61), andic properties were used to explain high SOC stocks. I would skip “andic properties” here because the processes responsible for high SOC stocks are given.
Material and Methods: Estimation of andic properties: I would suggest to start with the most important criteria according to soil classification (e.g. Alo + 0.5 Feo)
Line 433-434: oxalate extractable Fe is not equal to ferrihydrite
I suggest to reduce the number of tables and figures of the main manuscript. Please transfer some information into the electronic appendix (e.g. table 2?, table 3?, ??). There is no need for 11 figures in the main manuscript.

Dear authors,
Thank you very much for the revision you did. You really improved the manuscript a lot and there are just some minor issues which need to be addressed. I added my comments directly to the pdf files of the manuscript and the supplement. I will send you the supplement in an extra email because the websystem of the journal does not allow to attach two files.
I look forward to see the final version of your paper.
Best wishes,
Karsten Kalbitz

Dear authors, it is a pleasure for me to inform you that your manuscript is accepted now for publication in SOIL.

Sincerely

Jorge
On Behalf of Executive Editors of SOIL