Picea farreri Page & Rushforth

Picea farreri was described in 1980, from Farrer 1435 gathered in the Feng-shui-ling valley near the Myanmar–Yunnan border. Accounts vary as to whether the collecting locality was in Myanmar or China, but the Edinburgh herbarium database, where cones from this collection are deposited, files this material under China (E00540618) and Farjon (1990) gives the coordinates 25° 30’ N, 98° 15’ E, which fall within China. Farrer made this collection in 1919, the year before his premature death aged just 40 (Grimshaw & Bayton 2009) and a tree raised from Farrer 1435, once growing at Exbury (Hampshire, UK) formed part of the type specimen. This was a remarkably beautiful tree with rather glaucous leaves, drooping branchlets, and a rather open habit recalling that of the earliest introductions of P. spinulosa from Sikkim; furthermore it was the only living tree from Farrer 1435 traced by Page & Rushforth when they described P. farreri in 1980 (Page & Rushforth 1980).

This tree was propagated by grafting in the Hillier Nurseries in the 1960s, and later by Keith Rushforth and others, such that this clone has become well established in gardens throughout the UK. Prior to 1980, it had been identified as P. spinulosa or sometimes P. brachytyla. The original tree had reached 18 m tall × 56 cm dbh by 1987 when an adjacent oak tree, which had been suppressing it, fell in the great storm of 15–16 October; this left the spruce quite exposed and it died a few years later, following the drought of 1995 (Tree Register 2024). A graft planted in the sheltered valley at Tregrehan, Cornwall in 1994, was 20 m × 37 cm thirty years later, clearly having grown rapidly from about 6 m in 2008 when seen by John Grimshaw during research for New Trees (Grimshaw & Bayton 2009). Another graft at drier Thenford, Northamptonshire, probably dating to about the same time, was 17 m × 34 cm in 2019. A 1966 graft from Hilliers, in the even drier and much warmer conditions of Cambridge University Botanic Garden was 8 m × 22 cm and in poor condition in 2022; a graft from Keith Rushforth, planted at Sandringham Norfolk, in 1996 was 6 m × 15 cm and ‘not really thriving’ in 2019 (Tree Register 2024). Clearly this is a tree that favours a mild, wet climate. A young graft at the Yorkshire Arboretum, planted in 2014, was 3.5 m × 9 cm and a picture of health in 2023, but whether it thrives long term even this far north on the drier east coast of England remains to be seen. The Yorkshire plant is the most northerly listed by the Tree Register of Britain and Ireland, but this material would surely thrive on the west coast as far north as Inverewe in Scotland, and perhaps further inland in favourable sites, too, and across the island of Ireland. A tree at Forde Abbey, Dorset, 9.5 m × 36 cm in 2023 from a 1996 planting, is notable for having been raised from seed from the original Exbury tree rather than via grafting; seed raised plants of Farrer 1435 do appear, so far at least, to come true (Grimshaw & Bayton 2009).

Until about the time of the Exbury original’s demise, Picea farreri had generally been assumed to be a narrow endemic with a scattered distribution in the southern Gaoligong Shan (the mountain range rising to the west of the Salween River to form the frontier between China and Myanmar). From the mid 1990s, however, further collections would suggest a wider distirbution. Herbarium material collected during the Gaoligongshan Biotic Survey expeditions of 1996, 2000 and 2002 was later identified as comparable with P. farreri, including GLGS 16793 from Gongshan County, some 250 km further north from Farrer’s locus classicus (E00416057). This possible extension up the Gaoligong Shan would be concordant with local phytogeography and is to be expected from a wind-dispersed conifer; a wider east-west distribution at the northern end of its distribution, giving a pan-shaped distribution pattern with the Gaoligong Shan the panhandle, is also possible, and would explain the results of some studies, notably Shao et al. (2019) and Shen, Ran & Wang (2019) that each demonstrated a close relationship between P. farreri and certain spruce populations in northwest Yunnan previously attributed to P. brachytyla – q.v. for further discussion. Further research is needed, but the unique challenges of undertaking thorough fieldwork in this region should not be underestimated; the extraordinary topography, with the three famous parallel rivers and associated dividing ranges, means that locations just a few horizontal kilometres apart might take several days to travel between (Debreczy & Rácz 2011).

Nevertheless, the gradual introduction to cutlivation of additional material is bound to result in forms differing from the familiar, distinctive, almost-clonal population derived from Farrer 1435. This is to be welcomed, as Picea farreri is threatened in the wild (Zhang, Katsuki & Rushforth 2019) and broadening the cultivated gene pool is an important tool in the conservationists’ arsenal. Owen Johnson has already observed that MF 94–2116 at White House Farm, Kent, UK, originally introduced as P. brachytyla but considered by Keith Rushforth to represent P. farreri, differs markedly from Farrer 1435 in ‘Its neatly conic habit and spreading shoots’ (Tree Register 2024). Although grafts of the original Exbury tree seem to produce true seedlings, it cannot be taken for granted that they always will, and it seems prudent to provide a cultivar name for clonal material.