Earlier this year, a study published in the prestigious journal Science shook up the biology world by turning an accepted paradigm of plant growth on its head.

But now a pair of researchers is calling the findings into question, but the authors of the original work are standing their ground.

In comment pieces in the current issue of Science, the two sides face off over issues of soil conditions, plant biology and experimental design.

The original work, led by Peter Reich at the University of Minnesota, US, sought to understand how plant growth is affected by long-term high carbon dioxide levels. For 20 years, the team compared two groups of plants that employ slightly different methods of photosynthesis: the C3 pathway and the C4 pathway. (Read our report of the study here.)

Prevailing plant biology dogma states that C3 plants are more sensitive to atmospheric carbon dioxide levels than are C4 plants. Therefore, plants using the C3 pathway should produce more biomass as levels rise.

Much to the researchers’ surprise, the C3 plants grew well initially, but then lost their edge after 12 years. Instead, the C4 plants showed accelerated growth in the last eight years of the study, with increases in biomass outstripping the control plants by as much as 24%.

To explain the switch in growth rates, Reid and colleagues hypothesised that long-term high carbon dioxide levels triggered changes in soil microbes and nutrient cycling — changes that favoured C4 plant growth but hampered that of C3 plants.

The findings suggested that it may be difficult to predict with certainty just how much atmospheric carbon can be captured by plants in the future. As the effects of anthropogenic climate change continue to unspool, Reich cautioned, “We shouldn’t be as confident [that] we’re right about the ability of … ecosystems to save our hides.”

Julie Wolf and Lewis Ziska from America’s USDA Agricultural Research, however, don’t fully agree.

It’s too early to say that the C3-C4 growth paradigm is invalidated, they say, with the evidence pointing to an alternative — and decidedly less revolutionary — explanation.

“The pattern documented by Reich et al. can be explained by considering the natural history of the experimental plants and soil,” the pair write.

First, they explain that the topsoil at the experimental sites had been scraped away and treated with chemicals before the experiment began. The resulting sandy, well-drained conditions would ultimately favour C4 plant growth over longer timescales.

They then point out that species diversity was low within the experimental sites. With a maximum of four plant species in each plot (and all of them grasses), Wolf and Ziska believe the results should not be extrapolated to “make a broad statement about the general responses of C3 and C4 grasses to elevated CO2.”

Finally, they also question whether the experimental design and statistical analysis can support the conclusions drawn.

However, Reich and colleagues hit back, arguing that these criticisms are unfounded.

They acknowledge that the soil plots were indeed processed prior to the experiment, although in a different way than Wolf and Ziska describe. Nevertheless, they assert that the soil setting mirrors the disturbances observed in Earth’s grasslands due to grazing, cropping and altered fire regimes, and therefore remains relevant to the discussion.

Further, while physiology undoubtedly played a role in how the plants grew in the experimental habitats, the authors maintain that it is still unclear how these differences explain the observed responses to elevated carbon dioxide.

And the issue with statistics? The analyses used are robust to mixed sample sizes, although this was not explained in the original paper.

Putting their differences aside, both sets of authors agree that future research is needed to elucidate the mechanisms responsible for the switch in plant growth rates.

Ending their rebuttal on a conciliatory tone, Reich and colleagues come as close to waxing lyrical as is allowed in the pages of Science.

“Ecosystems change over time in complex ways that we are only beginning to understand,” the authors acknowledge. “Finding the appropriate context for field experiments is always challenging and should be done carefully.”