Types and other historical specimens of Allan and Richard Cunningham's taxa of Epilobium and Fuchsia ( Onagraceae ) from New Zealand in the Turczaninow Herbarium at the National Herbarium of Ukraine (KW)

. Thirteen historical specimens (including original material, mainly syntypes and isolectotypes) of Allan and Richard Cunningham's taxa of Epilobium and Fuchsia ( Onagraceae ) from New Zealand recently discovered in the Turczaninow Herbarium (KW-TURCZ) at the National Herbarium of Ukraine (KW) are discussed in comparison with digital images of and data on other relevant specimens from several other herbaria, such as K, BM, WELT, etc. These specimens, collected in New Zealand by Allan Cunningham in 1826 and 1838 and by Richard Cunningham in 1833–1834, represent the following taxa (as outlined by A. Cunningham): Epilobium alsinoides , E . atriplicifolium , E . confertum , E . hirtigerum , E . nerterioides , E . nummulariifolium , E . pedunculare , E . thymifolium , E . virgatum (A.Cunn., non Lam.), and Fuchsia procumbens . It is demonstrated that type designations for some names in Epilobium mentioned above remain problematic; some names appear to be not yet typified effectively. These problems are caused by several factors, such as misplaced original labels or specimens, errors or misinterpretations in curatorial or copied labels, a complicated numbering system applied by A. Cunningham, etc. Critical re-assessment of all available original and other associated specimens of the mentioned names in Epilobium , especially those in K, is needed. Before such re-assessment, we refrain from any nomenclatural actions affecting type designations. The recent discovery in KW-TURCZ of important historical specimens (not only Onagraceae but also many other plant families) collected in New Zealand in the first half of the 19 th century by Allan Cunningham, Richard Cunningham, J. Everard Home and some other early explorers is not only interesting from a historical viewpoint but also important for taxonomy, especially for proper application of names validated by A. Cunningham.


Introduction
Allan Cunningham (1791Cunningham ( -1839) ) was one of the most prominent figures in Australasian botany of the first half of the 19 th century (Heward, 1842a(Heward, , 1842b;;McMinn, 1970;Orchard, 2014;Heenan et al., 2017, etc.).His main botanical activities were focused geographically on Australia and New Zealand, but he also collected plants in Brazil (1814Brazil ( -1816)), Timor (1825; see Orchard, Orchard, 2013), and Mauritius (1821).Allan Cunningham visited the northern North Island of New Zealand in 1826 and 1838, basing himself in the Bay of Islands from where he collected numerous plant specimens.His brother Richard Cunningham (1793Cunningham ( -1835) ) also visited New Zealand and collected plants in 1833-1834 from a much wider area of the Northland Peninsula than his brother was able to achieve (Hooker, Cunningham, 1837;McMinn, 1970, etc.).Herbarium specimens of Allan and Richard Cunningham are known to be deposited in several herbaria.The main sets of their plants are reported to be in K, BM, and OXF, while smaller collections are known in CGE, E, G, NSW, PDD, W, WELT, etc. (Stafleu, Cowan, 1976;Mabberley, 1978;Orchard, 2014, etc.; herbarium acronyms are given according to Thiers, 2008-onward).Orchard (2014), who traced the distribution of Cunningham's specimens to various herbaria of the world, did not mention that the National Herbarium of Ukraine (KW; also known as the herbarium of the M.G.Kholodny Institute of Botany of the National Academy of Sciences of Ukraine, Kyiv) also holds important collections of the Cunningham brothers.Notably the KW herbarium is also not mentioned in other publications and online resources listing herbaria in which Cunningham's collections are known to be kept (e.g., Stafleu, Cowan, 1976;Thiers, 2008-onward, etc.).
Our preliminary inventory of selected taxa [for example, Veronica L. sensu lato (see Albach et al., 2018), some Geraniaceae (Mosyakin, de Lange, 2019), etc.] in the Turczaninow herbarium at KW demonstrated that it contains numerous specimens from New Zealand collected in the first half of the 19 th century, in particular, those of Allan Cunningham, Richard Cunningham (see biographic references above), and J. Everard Home (see Godley, 2010;Gardner, 2020;Mosyakin, de Lange, 2020), including many type specimens.Most probably these Cunningham's historical specimens from New Zealand were received by Turczaninow from the Royal Botanical Gardens Kew (K) through J.D. Hooker, together with other specimens from Australasia and some other parts of the world.It is documented that Turczaninow exchanged letters and specimens with Hooker (Kamelin, Sytin, 1997;Shiyan, 2011;Mosyakin et al., 2018;Mosyakin, de Lange, 2019, 2020, etc.).Further studies are needed to make an inventory of the This article is dedicated to Peter H. Raven (President Emeritus, Missouri Botanical Garden, St. Louis, Missouri, USA; Member of the National Academy of Sciences of the USA, Foreign Member of the National Academy of Sciences of Ukraine, etc.) in recognition of his fundamental contributions to the taxonomy and biogeography of Onagraceae, especially Australasian taxa of Epilobium New Zealand historical collections at KW, which will definitely result in identification of many other original specimens not yet known to the international botanical community (see Mosyakin, de Lange, 2020).
Here we provide the list of original and some other historical specimens of Cunningham's taxa of Epilobium L. and Fuchsia L. (Onagraceae) from New Zealand recently revealed in the Turczaninow historical herbarium at the National Herbarium of Ukraine (KW), with miscellaneous nomenclatural and taxonomic comments.In this publication we refrain from any nomenclatural actions (such as second-step lectotypifications or other corrected type designations) because we think that any changes to the nomenclatural status quo regarding the types of New Zealand names in Epilobium should be introduced (if indeed deemed necessary) only after a thorough additional examination of all relevant specimens available at K, BM, WELT, and other herbaria holding specimens collected in New Zealand by Allan Cunningham and Richard Cunningham.

Authorship of some names in Onagraceae validated by Allan Cunningham (1839)
Most of the taxa of Onagraceae validated by Allan Cunningham (1839) in the eleventh part of his Precursor… are cited with the authorship "A.Cunn."(see Schönberger et al., 2018Schönberger et al., , 2019;;IPNI, 2020-onward;POWO, 2020-onward, etc.), with the exception of the name E. nummulariifolium (cited as authored by "R.Cunn.ex A.Cunn."), Fuchsia procumbens (cited with "R.Cunn."),and two species names (E. pallidiflorum Sol. ex A.Cunn. and E. pendulum Sol. ex A.Cunn.) based on manuscripts of Solander and properly cited with the authorship "Sol.ex A.Cunn.".However, in the protologue Allan Cunningham, by adding the note "(R.C.Mss.)", clearly attributed the authorship of names and descriptions of three species of Onagraceae to his brother, Richard Cunningham.Following Art.46.2 and 46.3 of the ICN (Turland et al., 2018), the names Epilobium nummulariifolium, E. thymifolium, and Fuchsia procumbens should be cited with the following authorship: "R.Cunn."(as the author of the names and descriptions published by A.

Numbers associated with Allan Cunningham's specimens: an explanation
When citing types and other specimens of Allan Cunningham, many authors (for Epilobium, mainly Raven and Raven, 1976) often provided numbers after the collector names, which can be interpreted as collection numbers.However, that is not the case.The problem with the Cunningham specimen numbers and his rather confusing numbering system was specially discussed by Orchard (2014: 44-45).Here we provide his most important conclusions: "Most authors have treated these shipping numbers as equivalent to collection numbers… <…> However, a label of the kind 'Cunningham 47', as appears on many herbarium sheets, is not particularly informative.Particularly for low numbers, the same number will exist for multiple collections.<…> Cunningham did not place any enduring value on these shipping numbers, although he retained them in his own herbarium as an aide memoire of what he had sent to Aiton and Banks.Later in life, when despatching [sic] replicates to friends and colleagues, he renumbered specimens in each despatch.Other numbers became attached to Cunningham specimens subsequently.Herbarium specimens were 'collectables' in 19 th century polite society, and were traded among dilettantes.When these private collections were later given to institutions, or sold, the private numbers went with them, and are often confused with Cunningham's shipping numbers.Other specimens may now be found with two numbers, the original shipping number, and Cunningham's secondary shipping number".
Thus, the shipping numbers or other kinds of numbers associated with Cunningham's specimens should not be treated as the unique collection numbers or unique identifiers of specimens or gatherings.The numbers listed by Raven and Raven (1976) and some other authors for types of names in Epilobium are mainly the original species numbers used by Cunningham (1839) continuously in all parts of his Precursor… published in 1837-1840 (see the combined facsimile edition: Cunningham, 2017).
Representatives of Onagraceae in general and Epilobium in particular are considered among taxa of angiosperms most interesting for studies of various aspects of plant evolution and biogeography (Raven, 1972(Raven, , 1973(Raven, , 1976;;Seavey, Raven, 1977b;Connor, 1985;Raven, 1988;Hoch et al., 1993;Martin, 2003;Katinas et al., 2004;Wagner et al., 2007;Kumar et al., 2018, etc.).Quite naturally, taxa of Onagraceae inhabiting rather isolated islands of New Zealand provide intriguing cases of diversification in isolation, following longdistance dispersal events.Correct taxonomic and phylogenetic interpretations of morphological and geographical patterns observed in that group are crucial for reconstructing its patterns of evolution and historical biogeography.And, since the application of plant names is achieved by their nomenclatural types (Principle II and Art.7.1 of the ICN: Turland et al., 2018), reliable taxonomy is impossible without linking the names to their types through proper typification.
The first fossil records of Epilobium pollen in New Zealand are probably dated by the (late?)Oligocene (Daghlian et al., 1984;Martin, 2003) though Mildenhall (1980) records the definite presence of Epilobium pollen from the Pliocene and that of Fuchsia from the upper Oligocene.The migration routes of Epilobium to New Zealand and biogeographic affinities of native New Zealand taxa of the genus remain debatable, with possibilities of both Eurasian relationships (Raven, 1988) and, for some taxa, hypothetic migrations from/to South America, either due to long-distance dispersal events or via the ancient Antarctic route.However, available evidence (taxonomy, karyology, etc.) overwhelmingly indicates that ancestors of Australasian taxa of Epilobium most probably migrated to Australia from Eurasia, and thus most of New Zealand species had their direct ancestors in Australia (Raven, 1972(Raven, , 1973;;Raven, Raven, 1976;West, Raven, 1976;Hair et al., 1977;Seavey, Raven, 1977b).It is noteworthy that Seavey and Raven (1977a) demonstrated that South American taxa of Epilobium have been derived from at least two separate introductions from North America and two from Australasia (Australia or/and New Zealand?).Unfortunately, molecular phylogenetic data on Epilobium, and on its Australasian taxa in particular, remain rather limited (see Baum et al., 1994;Levin et al., 2004;Lorimer, 2007, and references therein).Further details of the complicated evolution of Epilobium in time and space will become available after additional studies.
Several species of Epilobium native to New Zealand are known as introduced and naturalized in Europe, especially the British Isles and some continental European areas (Davey, 1953(Davey, , 1961;;Holub, 1978;Jørgensen, 1992;Stace, 2010;Pyšek et al., 2012;Kaplan et al., 2018, etc.).The taxonomic identity of several alien taxa in Europe was rather confusing in the past; some of them were known under misapplied names.Their hybrids with other species were reported as well (Kitchener, McKean, 1998;Stace, 2010), which further complicated their identification.In the recent edition of the New Flora of the British Isles, Stace (2010: 361) Rosati et al., 2020) and E. melanocaulon Hook.found at the Ziller River, Mayrhofen, Austria (de Lange, unpubl.data, P.J. de Lange OST1, AK342661).Thus, proper typification of names will ensure the proper taxonomy and nomenclature not only for species of Epilobium native in New Zealand but also for introduced and potentially invasive plants occurring beyond their native ranges.
Notes: We were unable to find any unquestionable original specimens of E. alsinoides in the online databases of the JSTOR Global Plants and the Royal Botanic Gardens, Kew.The two sheets at K, K000742561 and K000742562, containing plants currently identified as E. alsinoides subsp.alsinoides, are in fact considered to be original specimens of E. thymifolium.Both these specimens have the labels indicating their locality as "About the Waimaté Station | among ferns", and both are dated by 1834, not 1833 (see discussion below under E. thymifolium).Thus, judging from available data, these two specimens from K cannot be regarded as unquestionably belonging to original material of E. alsinoides.Raven and Raven (1976) mentioned a specimen from WELT as an isotype (isolectotype?) of E. alsinoides.We were able to find online data for a specimen with the WELT Registration Number SP079421 identified as E. alsinoides, with its type status indicated as "uncertain" (data available from https://collections.tepapa.govt.nz/object/719249; no image was available when the online resource was accessed on 14 February and 2 March 2020).Judging from online data, this specimen was collected by Richard Cunningham in 1834 (January?), and its location was indicated as "New Zealand", with additional information in square brackets, which was most probably partly derived from the protologue: "[New Zealand, North of North Island, probably Bay of Islands, Hokianga or Whangaroa Harbour area]".Judging from the available data, this specimen does not belong to syntypes of E. alsinoides but may be part of original material (as defined in Art 9.4 of the ICN: Turland et al., 2018).
In order to solve the problem of the proper typification of the name E. alsinoides, additional searches for any physical original specimen( s) not yet available online should be performed, first of all at K and WELT.If no unquestionable original specimens corresponding to the protologue (syntypes) are available in K, the KW specimen KW001003054 will be eligible for lectotype designation because it perfectly matches the protologue and is thus a syntype.It should be noted that, according to Art. 9.12 of the ICN (Turland et al., 2018), in lectotype designation a syntype must be preferred over paratypes (if such exist) and/or the uncited specimens and cited and uncited illustrations that comprise the remaining original material (if such exist).A.Cunn.,Ann. Nat. Hist. 3(14)  Current taxonomic status: Raven and Engelhorn (1971; see also Raven, Raven, 1976) considered this taxon as a subspecies of E. alsinoides.This concept is accepted by Schönberger et al. (2018Schönberger et al. ( , 2019) ) and in POWO (2020onward), but not by the New Zealand Plant Conservation Network (www.nzpcn.org.nz,accessed 2 March 2020).

Epilobium atriplicifolium
Notes: There is only one known original specimen at K (K000742563) which is annotated as E. atriplicifolium and labeled as the "holotype", with the following locality information on the label: "Towards the Great falls Keri Keri".We accept that specimen as the lectotype designated by Raven and Raven (1976: 190), despite some difference in label data.The original specimen at KW (KW001003057) is thus accepted as an isolectotype; however, it should be noted that the label of the KW specimen is a better match to the protologue data.A.Cunn.,Ann. Nat. Hist. 3(14): 34.1839.
Note: We were unable to find online data on and images of any original specimen(s) of E. confertum in the JSTOR Global Plants (https://plants.jstor.org)and the Kew Herbarium Catalogue (http://apps.kew.org/herbcat/gotoHomePage.do).A search in K for any extant original specimen(s) of E. confertum should be attempted.If there is/are no such specimen(s) present at K (lost or destroyed?), the KW specimen (and other original specimens in JE and WELT reported by Raven and Raven in 1976?) will be available for lectotype designation to replace the presumably lost or destroyed lectotype from K designated earlier by Raven and Raven (1976).A.Cunn.,Ann. Nat. Hist. 3(14): 33.1839.
Notes: The K specimen currently available from online resources of JSTOR Global Plants and the Kew Herbarium, which we consider at present to be the standing lectotype (K000742597, image available from https://plants.jstor.org/stable/10.5555/al.ap.specimen.k000742597 and http://specimens.kew.org/herbarium/K000742597), has five labels and notes.One label indicates that it is E. cinereum A.Rich. and lists the number 544 corresponding to the number of that species in the Precursor… (Cunningham, 1839).However, R. Melville in 1959 added the following note (now mounted on the sheet in the bottom left corner): "The wording on Richard Cunningham's field label refers to A. Cunn.no.546 in Ann.Nat.Hist 3, 33 (1839).The No. 544 label is therefore misplaced as this is part of the type material of Epilobium hirtigerum A. Cunn.l.c.".We were unable to find a matching specimen in the online database of WELT (https://collections.tepapa.govt.nz).
Moreover, there is definitely some confusion with the dates.The specimens K000742597 and KW001003059 bear the collection year 1834 (not 1833 cited in the protologue!).Thus, these K and KW specimens cannot be formally considered as syntypes, unless their labels were erroneously dated, or unless there was an error or typo in the protologue.However, they are most probably parts of original material as defined by Art 9.4 of the ICN (Turland et al., 2018) and represent uncited specimens that were available to the author prior to, or at the time of, preparation of the description validating the name.Besides, the KW specimen (definitely with an original label) is much better preserved than that the small fragment from K.
To add to the confusion, the plant fragments currently mounted on sheets in Kew (K000742597) and Kyiv (KW001003059) are morphologically different and, in our opinion, may belong to two different species.Because the Kew specimen was probably mislabeled and the KW specimen still retains its original label, we think that KW001003059 better represents the original concept of the species as intended by Allan Cunningham.This, it is concluded that the Cunningham's specimens of E. hirtigerum available at K and KW are not syntypes because of the labels dated by 1834, not 1833.If no syntypes dated by 1833 are available (for example, in WELT), then the type indicated by Raven and Raven (1976; we suppose it is K000742597) should be considered the standing lectotype, but only if we are sure that this specimen from Kew indeed matches the original concept of the species.We further conclude that at present the proper application of that name remains uncertain.A.Cunn.,Ann. Nat. Hist. 3(14): 32.1839.
Notes: The story of nomenclatural confusion regarding the name E. nerterioides and some other names applied to this and related species was presented by Melville (1960), who characterized the situation as the "chapter of errors" (l.c.: 297).
The K specimen K000742539 (image: https://plants.jstor.org/stable/10.5555/al.ap.specimen.k000742539and http://specimens.kew.org/herbarium/K000742539)contains three plant fragments.Most probably these fragments were mentioned by Allan (1961: 261) as "3 imperfect specimens at K that very doubtfully belong here" (i.e., probably do not belong to E. nerterioides).However, it seems that Melville (1960: 298) had no doubts regarding the identity of these three fragments, judging from his statement that "Three fragments of Allan Cunningham's material of E. nerterioides are preserved at Kew".
There An additional comment on the type status of BM000797681 appears as an anonymous handwritten note mounted below the BM000797681 plant: "Probably the "missing" specimen mentioned by Melville in Kew Bull.: 14: 298 (1960)".This note refers to the following text in Melville (1960: 298): "When, at the start of this investigation, the collections in the Herbarium of the British Museum were consulted, a much better and more complete specimen was seen bearing Allan Cunningham's original field label.It was evident that this specimen should be regarded as the holotype and the attention of the Museum staff was drawn to this fact.The sheet was put aside with the intention of placing it in a type cover and making a minor repair.Later, when it was desired to make detailed notes and comparisons, the sheet could not be found and now, after the lapse of more than a year, it is still mislaid.At present, the only specimen available on which to base a decision is the Kew isotype".However, no "Allan Cunningham's original field label" mentioned by Melville is currently attached to the specimen.
After studying the Kew specimen (K000742539) Melville concluded that "The name E. nerterioides A. Cunn.therefore belongs to the species that hitherto has generally been called 'E. pedunculare ', i.e nerterioides sensu Hook.f., and E. pedunculare sensu Hausskn., non A. Cunn.f. aprica Hausskn.).Judging from our comparison of the KW specimen KW001003056 with digital images of the mentioned specimens from K and BM, the Kew specimen K000742539 is a better match.Thus, it seems to confirm the conclusion of Melville (1960) regarding the former misapplication of the name E. nerterioides but not regarding his new species name E. inornatum, which is now considered a synonym of E. komarovianum H.Lév. (see Schönberger et al., 2018Schönberger et al., , 2019;;POWO, 2020-onward, etc.).
In our opinion, the KW specimen KW001003056 is not part of original material because its label does not match the protologue; however, considering the confusion around the application of the name E. nerterioides (see above), that specimen is important for fixing the identity of the species as originally understood by A. Cunningham.
The species epithet was originally published by Cunningham (l.c.) as "nerterioides" and that spelling was and still is followed in many publications and databases (IPNI, 2020-onward; POWO, 2020-onward); we also use the original spelling here.However, some authors think that the original spelling should be corrected to nerteroides, following Art.60.10 of the ICN (Turland et al., 2018) because the species was compared to a Nertera Gaertn.The corrected spelling is accepted in Schönberger et al. (2018Schönberger et al. ( , 2019)).
Original material (according to the protologue): "New Zealand (Northern Island).-1769, Sir Jos.Banks.Shores of the Keri Keri river, and in dry as well as in boggy grounds.-1834, Rich.Cunningham".
Notes: The presumable type specimen (lectotype) at K (K000742538; images: https://plants.jstor Plants to be an isotype (isolectotype?), and is marked as "Type" (holotype?) in the Kew herbarium catalogue (both databases accessed 16 February 2020).The specimen was presented to K by Heward (see the label information below; also Orchard, 2014) and is mounted on the same sheet with K000742537 (not belonging to original material).The following label is attached: "535 | Epilobium nummularifolium RC | New Zealand | A. Cunningham 1838 | Heward 1840".This specimen, as well as the two KW specimens reported above, was collected not by Richard Cunningham, but by Allan Cunningham during his second visit to New Zealand in 1838.Thus, K000742536 is not a syntype but it most probably belongs to original material of E. nummularifolium because we can justly assume that it was available to A. Cunningham prior to preparation of the description validating the name.
The original specimen from K has no label corresponding to the locality information given in Allan (1961) and Raven and Raven (1976).Despite that, since the second specimen in K K000742536 does not belong to syntypes, we think that no second-step lectotypification is needed, and thus we accept the specimen K000742538 (in fact, the only one reliably known original specimen at K matching the protologue of E. nummularifolium) as the standing lectotype designated by Raven and Raven (1976).
There are at least two other original specimens at BM, a presumed isolectotype BM000797684 (image: https://plants.jstor.org/stable/10.5555/al.ap.specimen.bm000797684) and the syntype BM000797683 (image: https://plants.jstor.org/stable/10.5555/al.ap.specimen.bm000797683, the specimen containing three plants collected by J. Banks and D.C. Solander).The specimen collected by Banks and Solander is definitely part of original material; the second specimen, BM000797684, does not belong to syntypes.Judging from handwriting and specimen's metadata, the plants mounted on BM000797684 were collected by Allan (not Richard!) Cunningham.They are undated, but the year of collection 1838 cannot be excluded.If so, BM000797684 and the two specimens in KW are probably duplicates.
An additional (the third?) presumably original specimen of E. nummulariifolium is present in BM; it is annotated by P.H. Raven and T. Engelhorn as a probable isotype (BM000797680: https://plants.jstor.org/stable/10.5555/al.ap.specimen.bm000797680).The specimen is mounted on the same sheet with BM000797681 representing E. nerterioides (see above).The label of BM000797680 seems to be that of Allan (not Richard!) Cunningham, and the specimen is undated, so its status as part of original material is questionable.
An original specimen from WELT mentioned by Raven and Raven (1976) probably corresponds to the specimen with the WELT Registration Number SP079420 collected by Richard Cunningham in 1834 (January?)and currently identified as E. nummulariifolium var.angustum Cheeseman (data available from https:// collections.tepapa.govt.nz/object/719248;no image was available online on 14 February 2020).
All three historical specimens from KW (see above) were originally held in the same Turczaninow's species folder labeled as "Epilobium nummularifolium".Judging from the labels (including the year 1838 and handwriting), the specimens KW001003060 and KW001003061 were collected by Allan Cunningham during his second visit to New Zealand (not by Richard Cunningham in 1834); thus they are not syntypes but anyway they belong to original material of E. nummulariifolium (uncited specimens that were available to the author prior to, or at the time of, preparation of the description validating the name).The label of KW001003062 is undated and it is not written by any of Cunningham brothers.Judging from similar curatorial labels associated with other New Zealand specimens in KW-TURCZ, these plants were probably collected by J. Everard Home in the 1840s, and thus the specimen KW001003062 is not part of original material.Cunningham (1839: 31) cited "E.pendulum.Sol.Mss. in Bibl.Banks" in synonymy of his E. nummulariifolium.Consequently, the name E. pendulum has not been validly published (Art.36.1(b) of the ICN: Turland et al., 2019).A.Cunn., Ann. Nat. Hist. 3(14): 31. 1839.
Notes: The two K specimens currently listed as E. pedunculare in the JSTOR Global Plants online database, K000742583 and K000742584 mounted on one sheet (https://plants.jstor.org/stable/viewer/10.5555/al.ap.specimen.k000742583and https://plants.jstor.org/stable/10.5555/al.ap.specimen.k000742584,respectively; also available from Kew at http://specimens.kew.org/herbarium/K000742583 and http://specimens.kew.org/herbarium/K000742584, respectively), do not belong to original material of E. pedunculare.They are both dated by 1840 and properly refer to original material of the name E. linnaeoides Hook.f., which is now considered to be a synonym of E. pedunculare (see Melville, 1960).
However, K000742582 (https://plants.jstor.org/stable/10.5555/al.ap.specimen.k000742582and http:// specimens.kew.org/herbarium/K000742582)cited in JSTOR Global Plants as the type of Epilobium caespitosum Hausskn.(also considered a synonym of E. pedunculare; see Melville, 1960), seems to be the real type of E. pedunculare.It is dated by 1834 and was collected by Richard Cunningham "between | Waimaté | & the great forest | of Hokianga | past [east?] the lake | New Zealand", which quite closely matches the protologue.We accept this specimen as the standing lectotype that first Allan (1961) and then Raven and Raven (1976) had in mind.
It is interesting that the KW specimen is clearly dated as 1833 (not 1834 as indicated in the protologue) and thus its type status can be questioned.Alternatively, probably there was an error in labeling.Because of that we accept here KW001003063 only as a possible isolectotype.R.Cunn. in A.Cunn., Ann. Nat. Hist. 3(14): 32. 1839.
Notes: Allan (1961: 267) provided the following type information: "Type locality: Among fern in dry exposed situations at the Waimate Mission Station.Type: R. Cunningham, 1833".He also stated that "The material of R. Cunningham at K, consists of 4 poor scraps without capsules.I did not find the BM material".That statement does not constitute even a partial typification because, in our opinion, Allan simply indicated that he examined some possible syntype or syntypes at K. The "4 poor scraps without capsules" mentioned by Allan probably refer to the specimen K000742561 (see below).Raven and Raven (1976: 198)  In contrast to the K specimens mentioned above, the KW specimen (KW001003055) is dated by 1833, which matches the protologue.
There are at least two possible explanations.According to one, an incorrect date was cited either in the protologue (1833 instead of 1834?) or on the labels of the two Kew specimens (1834 instead of 1833?).However, we now know that the specimen with the collection year matching the year indicated in the protologue does exist, and it is KW001003055.Because of that we may assume that Richard Cunningham probably collected E. thymifolium both in 1833 and 1834.In that case the specimens collected in 1834 are not syntypes but they are still parts of original material (uncited specimens most probably available to the validating author before his publication of the name and description).If it is indeed the case, then the lectotypification by Raven and Raven (1976: 198) may be superseded because in lectotype designation a syntype should take precedence over uncited original specimens (Art.9.12 of the ICN: Turland et al., 2018).If that viewpoint is accepted, it might be suitable to designate the specimen KW001003055 as the lectotype of the name E. thymifolium because that specimen matches the collection year (1833) and other data indicated in the protologue and is thus a syntype.However, we refrain from that nomenclatural action because of the reasons stated above in the Introduction, in particular, because of the need of a thorough re-examination of original specimens of Epilobium at K, BM, WELT and other herbaria where New Zealand specimens of Allan Cunningham and Richard Cunningham are or may be present.
It is also important to emphasize that plant fragments currently mounted on the Kew specimen K000742561 seem to be morphologically non-uniform.As far as we can judge from studying the high-resolution digital image of K000742561 (see the links above) and physical specimens in Kyiv (KW001003054 and KW001003055, see Figure 1), only the uppermost fragment on K000742561 (mounted directly below the barcode) matches morphologically the two plant fragments available on KW001003055 (E.thymifolium).All other fragments on K000742561 seem to be corresponding morphologically to KW001003054 (E.alsinoides s. str.).Thus, it is quite possible that the sheet K000742561 either represents a mixed collection or contains incorrectly mounted specimens in fact corresponding to two species, E. thymifolium and E. alsinoides, as originally understood by A. Cunningham.A.Cunn., Ann. Nat. Hist. 3(14): 33. 1839, nom. illeg. (non Lam., Encycl. 2(1): 375. 1786).

Epilobium virgatum
Original material (according to the protologue): "New Zealand (Northern Island).In woods near the falls of the Keri-Keri river, at the head of the Kana-Kana river, &c., Bay of Islands.-1834.R. Cunningham".
KW specimen: "Epilobium virgatum RC. | In woods near the falls of the Keri-Keri R. | Bay of Isl ds | N. Zeal d | 1834 RC.", KW001003065.Syntype.It should be noted that the label of the KW specimen better matches the protologue than the label of the lectotype from K, as reported by Raven and Raven (1976: 120).
Current taxonomic status: illegitimate name (later homonym); now considered to be a synonym of E. billardiereanum DC. subsp.cinereum (A.Rich.)P.H. Raven & Engelhorn (1971: 349;see Schönberger et al., 2018see Schönberger et al., , 2019)).In our opinion, any future researchers engaged in further taxonomic and phylogenetic studies of New Zealand taxa of Epilobium should start with critical reassessment of all available types and other original and historical specimens of Allan and Richard Cunningham (as well as other prominent collectors), in particular, those deposited in K, BM, WELT, and now also in KW.The specimens from KW-TURCZ are especially important in that respect because they preserve the original labels, which have not been replaced by curatorial ones, and which are reliably associated with their corresponding specimens (not misplaced and mislabeled, as it was reported for some specimens at K, see above).
Original material (according to the protologue): "Totera indigenis.New Zealand (Northern Island).Around the village of Matauri on the east coast opposite the Cavallos Isles, inhabiting the sands immediately above the range of the tide, where it was found in flower in March.-1834,R. Cunningham".
This, an uncommon coastal species (de Lange et al., 2018), is known only from small populations along the far northern coast of the North Island (Northland and Coromandel Peninsulas), and from Aotea (Great Barrier Island) of New Zealand.The species is currently placed in the monotypic Fuchsia sect.Procumbentes E.J.Godley & P.E.Berry (Godley, Berry, 1995;Godley, Reynolds, 1998).It belongs to the well-outlined South Pacific clade (sections Skinnera (J.R.Forst.& G.Forst.) DC. and Procumbentes) confined almost exclusively to New Zealand and adjacent islands (with only one species endemic to Tahiti, Society Islands) and containing the following currently accepted species: F. excorticata (J.R.Forst.& G.Forst.) L.f., F. ×colensoi Hook.f,F. perscandens Cockayne & Allan, F. cytrandroides J.W.Moore (Tahiti) of sect.Skinnera, and F. procumbens of sect.Procumbentes (Crisci, Berry, 1990;Berry et al., 2004).Martin (2003) analyzed the fossil record (mainly pollen) of Onagraceae in Australia and partly New Zealand and concluded that New Zealand representatives of Fuchsia may have been derived from some now extinct Australian taxa of the genus.The first fossil records of Fuchsia in New Zealand are dated by the late Oligocene (Mildenhall, 1980).A new fossil species, Fuchsia antiqua D.E.Lee, Conran, Bannister, U.Kaulfuss & Mildenh., was described recently based on material from a drill hole north of Invercargill, Southland (Murihiku in Te Reo Māori), South Island of New Zealand.That species is dated by the Early Miocene and is associated with Fuchsia sect.Skinnera (Lee et al., 2013).Godley and Berry (1995: 502) cited the type of F. procumbens as follows: "TYPE: New Zealand.North Island: around the village of Matauri, on the E coast opposite the Cavallos Isles, inhabiting the sands immediately above the range of the tide, Mar.1834, (female), Richard Cunningham (holotype, K)".However, in view of the existence of the second original specimen recently found at KW (see above), the mention of the original specimen from K as the type by Allan (1961: 281), who cited the type locality from the protologue and added "Type: K, R. Cunningham.There are 5 small pieces with ♀", constitutes effective lectotypification.Thus, the K specimen should be corrected to lectotype, following Art.9.10 of the ICN (Turland et al., 2018).The KW specimen is thus an isolectotype.We were unable to find other original specimens of that species in online databases of other herbaria, which emphasizes the importance of the well-preserved original specimen from KW.

Concluding remarks
Here we report and discuss 12 original and some other historical specimens (mainly syntypes, some isolectotypes) of nine species-rank names in Epilobium and one original specimen (isolectotype) of one species of Fuchsia, all taxa published by described by Allan Cunningham and Richard Cunningham from New Zealand.
The fundamental taxonomic treatment of Australasian Epilobium by Raven and Raven (1976) still remains the most important and comprehensive foundation for our understanding of the taxonomy, biogeography and evolution of the genus in New Zealand.However, despite much research and the published taxonomic and floristic treatments available, there are still problems with the proper designation of types for names of several New Zealand taxa of Epilobium, and the type status of some specimens has to be clarified, for which critical reassessment of all available original and other associated specimens, especially those in K, is needed.Before such re-assessment, we refrain from any nomenclatural actions affecting the existing (or even de facto missing) type designations.
The recent discovery of important historical specimens at KW (in the Turczaninow historical herbarium, KW-TURCZ) collected in New Zealand in the first half of the 19 th century by Allan Cunningham, Richard Cunningham, J. Everard Home (see the references above), and also by some other early explorers, is not only interesting from a historical viewpoint but is also important for taxonomy, especially for proper matching of names validated by Allan Cunningham and their corresponding original specimens.The problems with typification of some Cunningham's names in Epilobium mentioned above (and also some other names of other taxa described by that author) are caused by several factors, such as lost of misplaced original labels, errors or misinterpretations in curatorial or secondary (copied) labels, a complicated and potentially confusing numbering system applied by Allan Cunningham, etc. (see further details in Orchard, 2014).
It has been also demonstrated that the printed text of Florae Insularum Novae Zelandiae Precursor "contains a large number of errors, which can be corrected using an autograph manuscript that is still extant" (Earp, 2016: 366).Thus, if some typographical or other errors are suspected in the protologues of taxa of Epilobium (for example, collection years or toponyms not matching the relevant data from the labels of available specimens, see above), it would be advisable to consult the manuscript still held by the Royal Botanic Gardens Archives, Kew, or its microfilmed copies (Earp, 2016: 368-369).
As far as we can judge from available data concerning not only Onagraceae but also some other selected families checked at KW, the KW-TURCZ specimens collected by Allan Cunningham and Richard Cunningham in New Zealand usually have original labels (not the curatorial or copied ones).Also, in some cases the KW material is more abundant and better preserved as compared to other known specimens of the Cunningham brothers available in other herbaria.
Special projects aimed at search, identification and digitization of historical specimen in the Turczaninow herbarium would be important, either based on a taxonomic approach (such as revisions of particular taxonomic groups) or on geographical regions.Both approaches were applied in the course of the three projects at KW supported by The Andrew W. Mellon Foundation during 2007-2016 within the framework of the African Plants Initiative, Latin American Plants Initiative, and the Global Plants Initiative.Data and images that resulted from these projects are available through the JSTOR Global Plants online resources (https://plants.jstor.org),but they still represent only a part of the vast number of types and other historical specimens held in KW-TURCZ.An international collaborative project aimed at searches for and revision of New Zealand (or all Australasian?)specimens in the Turczaninow memorial collection at KW would be very desirable.

Figure 4 .
Probable isolectotype?See the note below.

Figure 5 .
Figure 5. Fuchsia procumbens R.Cunn. in A.Cunn.KW001003053, isolectotype Cunningham; see Art. 46.3 of the ICN) or "R.Cunn. in A.Cunn." (if it is considered useful to emphasize that the names have been validated in the publication authored by A. Cunningham or if a bibliographic reference is cited, see Art. 46.2Note 2 of the ICN).
did not indicate any type material but merely stated that Hooker (1852) included E. confertum in a broad concept of E. junceum Sol.