Here is my expert synthesis based on the most current evidence:
The single most promising near-term path to field-deployable blight-resistant American chestnuts is recurrent genomic selection (RGS) as now practiced by The American Chestnut Foundation and its academic partners. This strategy is expected to produce a population of trees with sufficient disease resistance to survive in the wild within two breeding cycles. The landmark Westbrook et al. (2026) paper published in *Science* demonstrated that by sequencing genomes and comparing genetic patterns with real-world disease outcomes, resistance can be predicted using DNA data alone. Crucially, this approach can identify resistant individuals that retain the species' forest competitiveness: "With genome-enabled breeding, we expect the next generation of trees to have twice the average blight resistance of our current population, with an average of 75 percent American chestnut ancestry." The next generation of trees is expected to start producing large quantities of seed for forest restoration in the next decade. The older transgenic pathway via the Darling line has fractured institutionally — in 2023, TACF decided to withdraw support of the D58 transgenic chestnut petitions after observational data indicated inconsistent blight resistance, a negative impact on growth, and decreased survival rates, compounded by a lab error at SUNY-ESF that resulted in mislabeling between D54 and D58, which may have occurred as early as 2016 but was not discovered until 2023, meaning regulatory petitions were filed under the wrong identity. SUNY-ESF continues to pursue the Darling 54 line independently — USDA-APHIS completed its regulatory review and issued a preliminary finding that D54 is "unlikely to pose a plant pest risk," though other regulatory agencies, including the EPA and FDA, still have to complete their reviews. Nevertheless, the genomic selection approach offers a more robust and broadly supported route because it can generate genetically diverse, regionally adapted populations rather than relying on a single transgenic event.
The most significant recent scientific development is the February 2026 Westbrook et al. paper in *Science*, which integrated genomic, phenotypic, and reference-genome data at unprecedented scale. The study demonstrated that genomic selection integrated with continuation of hybrid breeding and phenotyping is likely to produce seed for restoration trials in 7–15 years. The team produced new high-quality reference genomes for both American and Chinese chestnut and showed that significant resistance gains are possible through selectively breeding trees with an average of 70 to 85% American chestnut ancestry. This matters because it resolved a long-standing uncertainty: the genetic architecture of blight resistance is polygenic and complex, meaning recurrent selection is likely to be more effective than backcrossing. Equally important, hybrids with around 70% American chestnut ancestry have substantial blight resistance but also show resistance to another problematic disease called root rot, caused by *Phytophthora cinnamomi*. Earlier simulation work had warned that root rot greatly reduced chestnut biomass on the landscape, even at the highest resistance levels observed, and that warming climate enhanced the virulence of the pathogen. The discovery that genomic selection can simultaneously capture resistance to both pathogens in a tree that retains forest-competitive growth is a genuine inflection point. Additionally, one of the most exciting findings is the demonstration of heritable blight resistance in a small number of pure American chestnut families, indicating selective breeding can improve blight resistance, though a broader genetic base would be required for restoration.
The biggest remaining bottlenecks are institutional, regulatory, and temporal. The transgenic pathway illustrates the difficulty: this is the first time a transgenic forest tree is being considered for restoration use, and the three-agency review (USDA-APHIS, EPA, FDA) has been underway for years with timing not predictable for any of the agencies. The split between TACF and SUNY-ESF over the Darling line — exacerbated by the revelation that SUNY-ESF researchers had engaged with private investors to commercialize the tree, contrary to regulatory filings and the understanding that the tree would be in the public commons — has fragmented political and fundraising capacity. On the science side, even with genomic selection, forest tree breeding faces irreducible generation times; chestnut trees typically require 5–7 years to produce pollen and female flowers in the field, though TACF is experimenting with accelerated light regimes that show pollen production in less than 2 years under 16 hours of high-intensity light. A further challenge is that restoration efforts will be more successful if targeted to latitudes, elevations, and site conditions where root rot is not expected to be present well into the future, including areas north of the historical chestnut range, meaning that the southern half of the species' former range — below roughly 40°N latitude where *P. cinnamomi* is widespread — remains extremely difficult to restore without combined blight and root rot resistance, which has yet to be achieved at operational scale. Field trial data also reveal context-dependent complexity: a 2026 silvicultural study found that smaller transgenic chestnuts had reduced survival compared to similarly sized non-transgenics, and transgenic growth rates slowed as trees got larger, underscoring that blight tolerance alone does not guarantee competitive fitness in forest settings.
New funding would have the highest impact-per-dollar if directed at three interconnected priorities. First, scaling up TACF's recurrent genomic selection program — genotyping, phenotyping, and controlled pollination — is rate-limiting. Genomic selection integrated with hybrid breeding is likely to produce restoration seed in 7–15 years, but this is only possible because of several decades of breeding and phenotyping that produced the required germplasm; accelerating genotyping throughput and expanding the breeding population would compress that timeline. Second, investment in speed-breeding infrastructure to shorten generation intervals would compound gains: high-intensity light chambers are already producing pollen in less than 2 years versus 5–7 in the field, helping select, improve, and scale better genetics much faster. Third, parallel investment in *Phytophthora cinnamomi* resistance screening is essential because alleles for resistance to root rot and blight are not linked, meaning a separate screening pipeline must be maintained and then resistance to both pathogens combined through intercrossing. The Forest Service and the U.S. Endowment for Forestry and Communities have framed the chestnut as an initial prototype whose approval would lay the framework for restoration of other tree species such as ash and hemlock. This multiplier effect means that dollars invested now in chestnut genomic selection, speed-breeding, and dual-pathogen resistance infrastructure establish institutional and scientific precedents far beyond a single species, making this one of the highest-leverage conservation investments available in temperate forest restoration today.