Quercus arkansana Sarg.

It is thought that Quercus arkansana is a relict species that formerly had a wider distribution and is now found only in scattered populations in the southeastern United States. E.J. Palmer (1925) proposed that it might be ‘an ancient species, probably once widely distributed over the Coastal Plain and now nearing extinction.’ In cultivation it remains a collector’s item, popular with specialists due in part to its threatened status, but it is not widely discussed in horticulture. It is nevertheless an attractive medium-sized tree, with somewhat unusual lobing, especially in leaves on the second flush of growth, where lobing is more distinct. The leaves resemble those of Black Jack Oak (Q. marilandica), but are thinner and do not have the yellowish hairs beneath typical of that species. In Q. arkansana leaves emerge covered with hairs but eventually become more or less smooth except for tufts of hairs in the axils of the veins (Heathcoat Amory 2009).

Arkansas Oak is endemic to the southeastern United States, overall uncommon and scattered over the Gulf Coastal Plain from south Georgia to eastern Texas, avoiding the Mississippi River Delta (Lance 2004). As its name suggests, it is most common in southwestern Arkansas, which represents the northwestern limit of the range. Its distribution is split into two disjunct areas on either side of the Mississippi River: to the west in southwestern Arkansas, western Louisiana and eastern Texas, and to the east in Alabama (which ranks second in abundance of the species), southwestern Georgia and the Florida Panhandle. It generally grows in the shaded understory of mesic pine woods as a medium-sized tree, midway up slopes and ravines above the heads of small creeks (Hunt 1986). Though it was historically more common across this range, it is now understood to be restricted to isolated populations where it occurs sporadically in stands containing from a few to a few dozen individuals. There are, however, areas of local abundance in southwestern Arkansas and the Florida panhandle, where healthy subpopulations support thousands of trees. In the western area of distribution it has been found in multiple degraded sites, and it is likely that it exists in similar unknown locations, where it is inconspicuous and has not yet been surveyed. Due to its fragmented distribution, sporadic occurrence, threats from commercial logging and projected loss of habitat due to climate change, it is categorised as Vulnerable on the IUCN Red List (Jerome, Wenzell & Kenny 2017).

Despite its origin, it is, according to Sternberg (2004), surprisingly hardy in cultivation, surviving in USDA Zone 5 in Illinois. On the other hand, Gert Fortgens reports that at Trompenburg, the Netherlands, it is not reliably hardy. He includes it among several species for which ‘promising young oaks may do well for a few years but inevitably a killing frost puts an end to them.’ They can be cultivated there only if grown in containers and overwintered in frost-free greenhouses (Chassé 2012).

It was introduced to Royal Botanic Gardens, Kew, from seed received from the Arnold Arboretum, the institution at which it was described, in 1912, the year following its publication. The tree survives at Kew, measuring 16 m × 43 cm dbh in 2022, but has been surpassed in size by other much younger trees in the UK. The tallest tree recorded in The Tree Register (2025) is at Sir Harold Hillier Gardens, Hampshire, planted in 1983 and 17.7 m tall with a 42-cm trunk diameter in 2023; the largest trunk diameter was measured on a tree at RHS Garden Wisley in Surrey, planted the same year and reaching 12.4 m × 62 cm in 2019. A handsome specimen grows in the gardens of Buckingham Palace, London, 10 m × 31 cm in 2021 (The Tree Register 2025). The oldest tree in cultivation, aside from the 1912 introduction, seems to be at Hackfalls Arboretum in New Zealand, where two trees have reached similar dimensions to the one at the Sir Harold Hillier Gardens: one was 17.3 m × 42 cm in 2020. They were both grown from seed obtained at Hackfalls, from a grafted tree received from the Hillier Nurseries, presumably in the late 60s or early 70s; the tree later died above the graft (Hackfalls Arboretum 2025).

Quercus arkansana is found in several oak collections in Europe, the oldest mostly accessioned in the 1990s or early 2000s. At Arboretum des Pouyouleix, France, a multistemmed specimen planted in 2004 was 11 m × 39 cm in 2021, and fruited for the first time that year (B. Chassé, pers. comm.). At nearby Arboretum du Passadou, a tree was around 15 m tall in 2019 (pers. obs. AJC). The oldest and largest Arkansas Oak in Europe, and likely the champion tree in cultivation, is at Bokrijk Arboretum in Belgium, from seed received from the Morton Arboretum in 1981, wild collected in Louisiana. It has grown extremely well, forming a well-structured tree with pyramidal habit, about 21.5 m tall and with a trunk 64 cm in diameter (pers. obs. RC 2025).

A tree at Grigadale Arboretum in Argentina was accessioned in 1997 and reached 11.9 cm × 22 cm in 2018. It was sourced, as were several of the trees in European oak collections, from Mallet Court Nursery (Grigadale Arboretum 2025).

Quercus arkansana was described by Charles Sargent in 1911 from specimens collected in Fulton, Hempstead Co., Arkansas, by B.F. Bush and Sargent. He described it as similar to Q. marilandica, from which it could be distinguished by the absence of persistent yellow hairs on the lower surface typical of that species, and to Q. nigra, which unlike Q. arkansana has leaves without stellate hairs. Sargent mentions a specimen collected by Mohr in 1880 in Conecuh Co., Alabama, which Mohr considered to be a possible hybrid of Q. nigra and Q. marilandica, but which Sargent understood was likely Q. arkansana. Trelease proposed that the taxon was indeed a hybrid, which he listed as Q. × arkansana in The American Oaks (1924), giving Q. marilandica and Q. nigra as the parents. E.J. Palmer in 1925 refuted Trelease’s proposal and defended the species status of Q. arkansana, noting that Arkansas Oak did not show the characteristics found in oak hybrids that easily betray them: instability of type and polymorphism in foliage and other parts, not only between individuals but also on different branches of the same plant, and a lack of uniformity and symmetry in the individual leaves. His defence of the taxon’s species status was couched in heraldic terms: ‘I am not willing,’ he protested, ‘to have the testimony of so important a witness as Quercus arkansana impeached or to see the bar sinister placed upon its escutcheon, since none of the facts seem to warrant such action.’

William Ashe described Quercus caput-rivuli in 1923 from Crestview, Okaloosa Co., in the Florida panhandle. His initial view was that it was possibly a hybrid, similar to Q. arkansana, from which it differed in its shorter petioles and the less prominent upper pair of lateral veins, rarely extending as awns. The following year, he revised his opinion and placed it as a variety of Q. arkansana, confirming it was not a hybrid. Having been able to compare the fruit of var. caput-rivuli to the type species, he found that Q. arkansana var. arkansana had acorns about fifty per cent larger than the var. caput-rivuli. He also reported new occurrences of the variety in Clay County, Georgia, and Pike Co., Alabama, suggesting a regional distribution in the southeastern limit of Arkansas Oak’s range. Palmer in 1925 was not convinced that it could be identified with Q. arkansana. This name is generally not recognised as a distinct taxon, which sadly robs us of one of the more unusual epithets in Quercus. There are few ‘double-barrelled’ epithets in accepted names of oak species, and this would be the only one not formed from personal names (the others are Q. john-tuckeri, Q. cornelius-mulleri and Q. guilielmi-treleasei). Ashe’s epithet is Latin, from caput (‘head’) + rivuli (‘small brooks or rivulets’). Though there is no explanation in the protologue, one can assume it refers to the oak’s occurrence at stream heads or heads of ravines (Hunt (1986) reports that the name describes its occurrence at the ‘head of a small brook’). No other plant has been named with this epithet, though many bear epithets based on caput-, especially the highly evocative caput-medusae (Medusa’s head). According to Beckman et al. (2021) Q. caput-rivuli may deserve species status, and further biosystematic examination should be carried out regarding the issue.