Basic reporting

Stuble et al. conducted a study that examined the effect of experimental warming on ant-mediated seed dispersal. They subjected ant communities at a southern site and at a northern site to experimental warming and measured seed dispersal in a plant species that is adapted to dispersal by ants. This is well-designed and well-presented study that makes a good contribution to understanding how climate change influences positive species interactions.

They manuscript is well-written and contains appropriate background information and references to the literature. The figures and tables are appropriate and the submission includes relevant and complete results. I only have a few minor suggestions (see general comments).

Experimental design

The research question is clearly defined and the design of the experiment is sound. My only suggestion (other than a few minor comments outlined the in the general comments section) is that more detail should be added to the methods section to explain what is known about the density and distribution of ant nests in the experimental chambers.

Consider breaking up the paragraph (L115-135) into two paragraphs. The first paragraph could describe the chambers and temperature treatments. The second paragraph could describe the distribution and density of ant nests and species in the chambers. This would allow for more room to explain this aspect of the design.

For example, the following details regarding the density and distribution of ants should be clarified: L118-L119 it is not clear here if this statement is referring to A. rudis only or all ant species (I assume the former). Was the estimation of 1 nest per m2 quantified by the authors? If so, the authors should provide statement that they produced this estimate and refer to this as unpublished data. If the authors did not quantify A. rudis nest density at their sites or in the chambers, then the authors could reference other studies that report high densities of A. rudis nests in forest ecosystems (Lubertazzi 2012 Psyche, also see references in Ness et al. 2009) and then say they assume similar densities at their sites (and in chambers). Also, the authors should add in some comments about the likelihood that nests of other ant species are situated within the chambers. I am assuming that there is much more variation in where ant nests of species (other than A. rudis) are situated relative to the experimental chambers. The authors make multiple statements in the text (L122, L191) that the treatments affect ants situated in the chambers. I am assuming that they are referring to A. rudis only and not all ant species, but as the methods are currently written this detail is unclear.

Validity of the findings

Minor comments in general comments section

Comments for the author

Here are a few minor suggestions:

L22 consider replacing “US” with “North America”, as this ecosystem extends into Canada; also in L64

L24 consider replacing “removal” with “dispersal”

L21 sometimes the mutualism is referred to as “ant-plant seed dispersal” and sometimes “ant plant seed-dispersal” throughout the manuscript

L62 this would be a good spot in the manuscript to define the term myrmecochory as ant-plant seed dispersal. Also consider using this term more often throughout the paper, instead of “ant-plant seed dispersal”.

L64 consider adding a line stating that myrmecochory is integral in shaping plant communities (e.g., Christian 2001, Nature)

L69 Myrmecochorous is spelled incorrectly

I think that it would be helpful to more specifically describe what a myrmecochorous plant is – i.e., plants that have seeds containing a fleshy lipid-rich appendage (elaiosome) that are attractive to ants

L70 consider replacing “single” with “a single or a few ant species”, as most systems seem to be dominated by keystone dispersers that are not a single species, but a group of species in a genera (e.g. Aphaenogaster (Ness et al. 2009), Rhytidoponera (Gove et al. 2007)).

L79 state southern and northern site here, given that these terms are used in the prediction statements

L106 consider moving this line up (i.e., as the second sentence in the methods). If not, consider removing the statement “near the southern extent and northern extent of several ant species” from L97.

Consider moving the statements describing the A. rudis complex in L112-114 up to L109 where the A. rudis complex is first mentioned. Alternatively, remove “(or at least a species in the A. rudis complex)” from L109 (as the A. rudis is not defined at this point).

L117 I don’t think that this statement is necessary

L119 consider replacing “moreover” with “in addition”; and “watched” with “observed”

L127 consider adding “For example in 2011, we found a… (unpublished data)”

L139 what were the seeds placed in or what is a cache (e.g. a petri dish, on the ground)? How big is a cache?

L142 consider adding a line stating that A. canadensis seeds are readily picked by A. rudis (e.g., Turner & Frederickson, 2013, PLOS one, and look through references in Table 2 in Ness et al. 2009)

L143-145 consider adding, “when seeds naturally dehisce in these locations”

L145 add the dimensions of the mesh cage and the mesh size

L151 were trials conducted on different days? If so, were the seeds stored in any particular way between collection and the trials given that elaiosome quality can degrade quickly once seeds are collected?

L288-L292 I think that this is an important point that could be discussed in more detail. Ant partner identity can be important in influencing the outcome of myrmecochorous interactions. Differences in body size, foraging behavior, or nest characteristics, for example, could determine the rate at which seeds are picked up, how far they are moved (as you mention), how they are processed in the nest, and where they are ultimately deposited (Giladi 2006 Oikos; e.g., Hughes & Westoby 1992 Ecology, Ness et al 2004 Ecology, Servigne & Detrain 2010 Ecological Research, Prior et al. in press Ecological Entomology). Further, some species picking up seeds (other than A. rudis) could act as low quality dispersers (e.g., damage seeds, or remove the elaiosome in situ without moving seeds) (Giladi 2006). These differences could scale up to have large effects on plant communities (Christian 2001). Thus, even though overall dispersal is not effected by temperature, alterations in which ant species move the seeds under different temperatures could potentially have large effects on plant communities.

Figure 2: error bars represent…