Late Permian wood-borings reveal an intricate network of ecological relationships

Beetles are the most diverse group of macroscopic organisms since the mid-Mesozoic. Much of beetle speciosity is attributable to myriad life habits, particularly diverse-feeding strategies involving interactions with plant substrates, such as wood. However, the life habits and early evolution of wood-boring beetles remain shrouded in mystery from a limited fossil record. Here we report new material from the upper Permian (Changhsingian Stage, ca. 254–252 million-years ago) of China documenting a microcosm of ecological associations involving a polyphagan wood-borer consuming cambial and wood tissues of the conifer Ningxiaites specialis. This earliest evidence for a component community of several trophically interacting taxa is frozen in time by exceptional preservation. The combination of an entry tunnel through bark, a cambium mother gallery, and up to 11 eggs placed in lateral niches—from which emerge multi-instar larval tunnels that consume cambium, wood and bark—is ecologically convergent with Early Cretaceous bark-beetle borings 120 million-years later.

Dear authors, this is an interesting and important manuscript on the early evolution of wood-boring beetles. I have some additional comments: Introduction: It would be important to start this section with a more general phrase on the earliest fossil record (body fossils as well as dates from phylogenetic reconstructions) of beetles, something like what you have written in line 168 to 171 in the supplement.
Results: The differentiation of five sequential phases of beetle activity is quite helpful for the reader. Unfortunately, these phases are not reflected in the figures or mentioned in the discussion in the main manuscript. Only in the supplemental text and Fig.S5 there is a connection to it. I would prefer to see a combined figure of Fig. 3 of the main text with figure S5 including an explanation of the main phases of beetle activity, as outlined in the beginning of the result paragraph.
Minor remarks: Line 118 (main text): I guess you mean "pairs" instead of "pars" of beetle larval mandibles.
Line 141 (main text): "lasting ten or more years" -in the supplement (figure caption of S5) you only mention 8 years.
Line 181 (main text): I would prefer "environments" instead of "environs" -but I´m not a native speaker, of course.
Reviewer #4 (Remarks to the Author): Wood borings have great potential to record in situ evidence for feeding behaviors of arthropods, which is among the most significant developments in the terrestrial ecosystem. Feng Zhuo and colleagues provide the earliest record of feeding trace, insect larvae and adults having been found in a fossil wood piece. They also make a nice and comprehensive summary of fossil wood borings and early polyphagan beetles. The study provides valuable insights into the early evolution of both insect-plant associations and beetles.
The paper is technically sound, with accurate descriptions. I don't have any objection to the details of the description. But the paper seems to be formatted in the Nature style. I suggest that the authors put the full descriptions of borings and their inclusions to the main text.
I have only two suggestions, which should be easily dealt with.
1 Modern wood-boring beetles (e.g. bark beetles) play a very important role in forest ecology, for example by creating complex early successional forest (e.g. Swanson et al., 2011, The forgotten stage of forest succession: early-successional ecosystems on forest sites. Frontiers in Ecology and the Environment). In my opinion, Permian wood-boring beetles probably have the similar role. I would suggest to provide a brief discussion about the potential significance of wood-boring beetles on the Permian forest ecology.

Responses to Editor and Referees
We thank the referees for the constructive comments, and respond to them on a point-by-point basis (in blue text), below.
Reviewer #1 Remarks to the Author (Anonymous) 6. General comments. "Investigations of fossil animal-plant-fungal interactions have greatly expanded in the last 15 years revealing a wealth of previously unexplored data elucidating the history of herbivory and saprotrophic guilds and webs of complex multi-organism interactions that are finally allowing palaeontologists to reconstruct energy flows within ancient ecosystems with confidence. Moreover, much of this data, preserved as damage on plant fossils, provides some of the only fossil evidence to understand the stratigraphic ranges of key arthropod taxa for which body fossils may be scarce or unknown. That said, there are still many gaps in our knowledge of the groups of organisms involved in ancient interactions, the developmental series of specific damage types and the patterns of tissue consumption through time. This study illustrates a remarkable example of damage to Late Permian conifer wood from China that is specifically attributable to wood-boring beetles. The results are unique for several reasons: 1. Most boring in Palaeozoic woods has been attributed previously to oribatid mites; 2. These are among the oldest confidently established beetle borings and certainly the oldest with complex tunnel geometry.
3. The damage in the Chinese woods shows successional development of the tunnel system apparently through several larval instars revealing partitioning of food resources and modifications to the feeding and digestive mechanisms through the animal's life cycle.
4. In contrast to almost all previous studies, this fossil example remarkably preserves fragments of the organisms involved in the damage to the wood (especially beetle and fungal parts), together with accessory biotic remains.
5. The fossil also reveals the host tree's response to damage via development of reaction tissues."

AUTHOR'S RESPONSE:
We agree and hope that this study presents an entirely new perspective to the establishment of a late Permian micro-community or organisms that inhabited a gallery-and-tunnel system in conifer wood.

Typographic errors and colloquial terms. "
The paper is worthy of publication in the journal, but there are a few minor points that need cleaning up and/or clarifying before final publication. These include: [1.1], there are several minor typographic errors scattered through the text that need fixing. (See comments of the attached pdf). There are also some colloquial terms in the text and supplementary information (such as 'spotty' and 'scraggly') that might be better replaced/clarified for an international audience."

AUTHOR'S RESPONSE:
We have implemented the suggested changes by Referee #1 in the attached PDF file. In the Main Text the colloquial "spotty" has been changed to "sparse".
(See Main Text new line 89) In the Supplementary Information, "scraggly" has been changed to "… and a jagged appearance …" (See Supplementary Information new lines 221-222) 8. Other late Paleozoic arthropod wood borers. "[1.2], it would be good to point out early in the introduction that other arthropods apart from beetles were extensively involved in wood boring by the late Paleozoic; and later in the text, explain briefly or use references to note how one would consistently distinguish beetle vs mite borings. Some of this information is currently in the Supplementary Data, but I think it would be useful to present this as a brief summary in the main text."

AUTHOR'S RESPONSE:
We have overhauled the introduction by inserting a new first paragraph that provides a broader context of Paleozoic wood borings and their arthropod fabricators. We broach the issue of distinguishing mite from beetle borings-a point that we discuss later in the manuscript. Our new introductory paragraph (Main Text new lines 76-88) is the following: Wood-boring is an iconic feeding behaviour among extant terrestrial arthropods, particularly cambium and wood feeding insects and detritivorous oribatid mites 1-3 . Arthropod wood borers feed in roots, twigs, stems and trunks of the dead or live woody plants, where they consume bark, phloem, sapwood and heartwood 1 . Although wood-boring has a long geological history traceable to the Late Devonian (ca. 382-358 million years ago, Ma), these occurrences are closely linked to the fossil wood record. Nearly all previous Palaeozoic records of wood-boring have been attributed principally to oribatid mites 3,4 . The miniscule tunnels of oribatid mites likely represent the oldest form of wood boring, and provide indirect evidence for stereotypical tunnels and their contained coprolite clusters that display multimodal, instar-related, size distributions 5,6 . However, some larger borings have been found in pith tissues of Pennsylvanian-age tree-fern axis6, which exhibit tunnel diameters approaching a tenfold increase compared to oribatid mites 3 , consistent with a beetle fabricator 4, and Supplementary Note 1 .
The following new Reference 5 has been linked to the new introductory paragraph above to better document the fossil record of Paleozoic wood-boring arthropods.  192-206 (2015).
Also see Point 24 of Referee #3 below. Fig. 2 are a little dark to see some of the details mentioned in the text (e.g., undigested tracheids). Could these images be lightened a little to show some more details of the coprolite contents without endowing them with excessive contrast?"

AUTHOR'S RESPONSE:
The images in the initial submission are of low-resolution according to the requirements of the Journal. In our re-submitted files images are of high-resolution, and the content of the coprolites are clearly visible.  Figures 1R and 1S. "[1.5] Fig. 1R and S are quite small to decipher the details of the structures being illustrated. Perhaps this is a consequence of shrinkage during the pdf-making process, but could these images be enlarged to provide more detail?"

AUTHOR'S RESPONSE:
The images in the initial submission are of low-resolution, based on the requirements of Nature Communications. In our current, re-submitted files the corresponding images are of higher-resolution, allowing detailed structures of the specimens to be clearly visible.

Full descriptions of borings moved to the main text. "
The paper is technically sound, with accurate descriptions. I don't have any objection to the details of the description. But the paper seems to be formatted in the Nature style. I suggest that the authors put the full descriptions of borings and their inclusions to the main text."

AUTHOR'S RESPONSE:
We have moved some of the material from the Supplementary Information into the Main Text, such as Fig. S5, which is now Fig. 3 AUTHOR'S RESPONSE: This is an important suggestion. Extant wood-boring beetles certainly play a very important role in shaping forest ecosystems. And they probably also played similar role in the Permian forest ecosystems. However, we do discuss the role of the potential significance of beetle's borings on the Permian forest ecology, its extirpation, and its reappearance during the mid-Mesozoic. This role is discussed in the following added sentence (Main Text new lines 255-258.) The demise of this unique wood-boring association among ancient conifers, fungi, beetles and other arthropods represent the beginnings of a food web centered in a closed, woody micro-environment that was soon extinguished but originated a new and expanded during the later Mesozoic 47 .