Michael Pratt's cover shot of a pink Caladenia alata on Journal 94 raised a few eyebrows. How did this far-north orchid get down to Rainbow Mountain? Michael found the more common white ones there too (pers. comm.)

The Column tried to explain C. alata's chequered past to new member Georgina Upson but kept stalling on minor difficulties which needed investigating. The inimitable sequence of events below is the result.

Michael's photo sparked memories in Colin Ogle who dug out his own Rainbow Mountain shots (J96:26) of C. alata from 15 Oct 66 first labelled C. carnea ssp. minor then C. catenata. The Column's 1960s shots of C. bartlettii had those names and a few others successively crossed out too; classification was such a shambles at that time. Peter de Lange's report of C. alata at Great Barrier Island and Waikumete Cemetery [J77:10] had been a big enough surprise but, south of Rotorua? Amazing! Or was it? Read on if you think you can handle another imbroglio.

Robert Brown started it all in his 1810 Prodromus [1] by first describing Caladenia alata R. Br. from Sydney, long before any Pakeha had reported it in NZ. The all important type specimen went to BM, the British Museum's herbarium.

JD Hooker would have made his own Caladenia collection in NZ in late 1841 [2] and he received specimens of NZ's C. alata (Fig. 1, below) and C. minor (Fig 2, below) from Edgerley of Hokianga and from Colenso. JDH's specimen sheet labelled "409 Caladenia N. Zealand" at K (Herbarium at Kew) had 13 specimens of open C. alata and 9 of C. minor in bud but his earlier Caladenia minor Hook. fil. sheet at K-L (Lindley's Herbarium at Kew) had 4 specimens of C. minor (pers. comm. B. Molloy). JDH did not recognise C. alata, either in his 1852 NZ Flora [3] or in his 1855 Tasmanian Flora [4] where his description confused C. alata with either Stegostyla angustata or S. gracilis.

This is evident by his own doubts, "I have referred this plant to Brown's C. alata with some hesitation;" and Fitch's drawing of the labellum with 4, not 2 rows of disc calli. Hooker's 1864 Handbook [5] surely had C. alata and others lumped into C. minor because his curious description [NZNOG J 84:5] including traits from several different taxa [J85:12], yet he was in a managerial role at Kew from 1855-1885 [2] and could have checked Brown's type specimens at the British Museum relatively easily.

Enter RH Matthews who, according to his letter of 5 December 1898, first sent a pickled specimen of Kaitaia C. alata to TF Cheeseman on 5 September 1898. No other Caladenia flowers this early so it had to be C. alata. Many more specimens were sent as a "new species" by RH. Matthews and his son HB Matthews according to their letters.

TFC also missed the Robert Brown connection. No wonder with Sir JD Hooker's erroneous description [4] so Cheeseman described C. alata as Caladenia minor  var. exigua Cheesem. in his 1906 Manual [6]. But he decided in 1913 [7] and in his 1925 Manual [8], published posthumously, that the Matthews' specimens were different enough from C. minor to have specific status as Caladenia exigua Cheesem. because of; earlier flowering, a midlobe with only one marginal callus instead of several, plus acute sepals and petals instead of obtuse.

What? TFC may have been looking at C. aff. pusilla (Fig. 3 below, also J92:13,17), a rare NZ Caladenia with obtuse petals or possibly C. bartlettii, (Fig.4 below) but C. exigua had acute petals.

That's okay, TFC had it right for his time, C. bartlettii wouldn't be split from C. minor Hook. fil. until 1949 [9] and C. aff. pusilla which was C. minor in the 2000 Field Guide, only got tagged out in 2002 by Ian St George [J82:15].

Czechoslovakian botanist Karel Domin, in 1915, started the C. carnea revolt by misclassifying
C. alata as Caladenia carnea R.Br. var alata (R.Br.) [10] for Aussie plants. His misclassification ran parallel with our C. exigua which held sway until 1944 when Australian Rev. HMR Rüpp [11] who must have read Domin's German language paper, saw reason to lump all the small Aussie and NZ Caladenia into Caladenia carnea R. Br. with numerous varieties but precedent would have ruled so he too missed the C. alata connection in NZ.

Thus in 1944, Rüpp reclassified ours anew as Caladenia carnea R. Br. var. exigua (Cheesem.) Rupp as reported by Dan Hatch in the 1945 Transactions [12]. Robert Brown (R. Br.) at least got into the citation even if for the wrong orchid. Rupp himself had doubts as he wrote to Frank Bartlett on 18 Nov 49 [J78:34], "It is rather curious that C. carnea in NZ, although showing the same tendency to develop distinct varieties, as it does in Australia, keeps them all very diminutive."

Dan was also uneasy with the Domin/Rüpp arrangement and stated [13], "Since then [1944], additional material of C. carnea var. pygmaea [RS Rogers's classification for C. minor and C. pusilla, J74:31] has raised some doubt as to whether it would not have been better to retain Hooker's name as a distinct variety. But even the varieties of C. carnea are themselves so liable to vary, that for the present, at all events, we think it best to let the matter rest." Seeds of dissent were emerging, it seems.

Rupp, a dedicated amateur orchidologist, got caught again (by Hooker's description? [4]) it seems, with a September 1953 specimen of C. alata from Wyong classifying it in 1954 as Caladenia holmesii Rüpp [14] but this was a purely Aussie sidetrack.

Moore and Edgar, in the 1970 Flora II [15], extended the Domin/Rupp lead and lumped all the NZ Caladenia, including our C. alata and Stegostyla atradenia, into Caladenia carnea R. Br. with no varieties. But they put a full page footnote in fine print, detailing the historic descriptions of the earlier varieties and species then being lumped but expressed their own doubts as, "No critical description of the type of the species has been seen, and a number of vars have been described in Australia where the limits of the species itself have been variously interpreted." Their page of fine print is now of far more value than their preceding description of misclassified "C. carnea".

Next, it seems, Dr. Winifred Curtis [16], stated a "comb. nov." in 1979 for our C. alata as Caladenia catenata (Sm.) Druce var. exigua (Cheeseman) W. M. Curtis - with Brown, its original describer, cut out of the citation but Smith, Druce, Cheeseman and herself all now included, despite none of them having any part in the classification of Caladenia alata R. Br. Curtis had broken the carnea deadlock but had gone from the frying pan into the fire by linking the whole genus to the related C. catenata, a large, white Australian species.

Don't weep, gentle reader, this is just getting interesting. You can be forgiven for thinking that successive taxonomists were getting further and further from the facts. They were none of them fools, far from it, these were the innovators following the basic precepts of science and building on the works - and errors, more's the pity - of their esteemed predecessors. However a vital link in the chain had been missed by all, Robert Brown had only his acutely brief description in Latin [1] plus his type specimen on the other side of the world at the British Museum so that, deciphering his original species, was an almost impossible logistical hurdle for Australasian students especially with Hooker's misinformation in [4] to misguide them.

Our Brian Molloy [17] said, in 1983, "According to the Australian botanist Don Blaxell, [18, 1980] C. catenata is the correct name for orchids grouped under C. carnea." On p19, Brian had John Johns' photos of three taxa of so called "C. catenata" which Brian clearly depicted as different. The Column identifies these 3 now as, Fig. 11, Caladenia aff. chlorostyla, Fig. 12, C. nothofageti and Fig. 13, C. aff. pusilla. C. alata would have been included within Brian's C. catenata of course but didn't feature here along with a number of other northern taxa unavailable to Johns' camera. Brian also mentioned Winifred Curtis' above contribution and wisely added, "We have not attempted to resolve these issues here." So Rüpp, Hatch, Molloy, Moore & Edgar, all had doubts about the C. carnea and C. catenata connection with NZ but were in no position to argue without access to BM & K specimens.

However, a light was beginning to dawn. Mark Clements, during one of his many visits to European herbaria, on a momentous assessment of Australasian orchids, examined Robert Brown's type material of C. alata on 1 Mar 84. [pers. comm. B. Molloy]. He also examined Hooker's type sheets at Kew for C. minor [19] but only noted, in March 1985, that the "4 specimens... do not represent typical C. carnea". But Mark looked further at BM & K-L and wrote to Brian on 19 Sep 87 announcing - wait for it - that C. exigua Cheesem was synonymous with Caladenia alata R. Br., finally, 177 years after R. Br.'s original description [1]. So, C. alata finally got recognition in New Zealand whilst C. exigua got quietly dumped.

Brian mentioned the synonymy at the Kaimaumau-Ahipara native orchid field days on 7 & 8 Nov 87, attended by Doug McCrae, Bruce Irwin et al. Bruce writes that Doug was pointing out C. exigua seed capsules on both those days but Doug first reported them as C. alata in Newsletter 24:10,11 Dec. 1987, quite casually, as he did Calochilus herbaceus, as though these two were well known. Not so, this revelation was a minor bombshell at the time.

David Jones, also reported in 1988 [20, p93] that "In New Zealand, this species (C. alata) has been known as C. exigua Cheeseman." so Mark's official 1989 announcement [21] of his incisive revelation, must have fallen a little flat after 2 other publications had stolen a key piece of his thunder. Nonetheless, Mark mentioned 5 earlier names for C. alata [21, p20] and one for C. minor [21, p28] where he designated one specimen as C. minor on Hooker's K-L type sheet and added that "All other syntypes are C. alata." 2 years after Doug had slipped C. alata into Newsletter 24.

Back on 4 Dec 87, Doug had visited Chris Ecroyd [pers. comm.] and spotted Chris's "Caladenia carnea" (NZFRI 12454) in the Forestry Research Institute Herbarium and re-determined this Whakarewarewa specimen of 29 Sep 82, as C. alata, drawing on his knowledge of one month back. Chris included this in a 1990 Bot. Soc Newsletter [22] but it missed the NOG Journal until Michael Pratt asked the question in Feb 2005.

But in September 2001 [23] David Jones, et al deduced that Caladenia flava (Pfitzer 1889) was the type for Caladenia so they chose a discarded historic epithet for the genus, as was their perceived duty and Australasian C. alata became Petalochilus alatus (D.L. Jones) D.L. Jones M.A. Clem.

Petalochilus? This was the genus that RS. Rogers had erected [24] for two species of HB Matthews from Kaitaia when TF. Cheeseman declared them freaks. As freaks, they did not get included in his 1925 Manual [8] which, coincidentally, had to be nearing publication at the time. But were they freaks? Petalochilus were undoubtedly Caladenia-like, possibly mutant C. minor having petaloid labella as in Thelymitra but HBM insisted, in letters to Cheeseman and to Rogers that extensive colonies were propagating by seed. That is the norm for Caladenia too [20, p90], so they were not sterile mutants. Why then shouldn't they be accorded specific and generic rank?

The tea tree habitat was subsequently cleared to pasture and this neglected genus was lost to Kaitaia. However, Bruce Irwin found a bud of Petalochilus saccatus at Scott Point on 29 Sep 97 [J65:14] so the genus might still be around despite Cheeseman and his NZ followers neglecting it into near oblivion. So it would seem inappropriate to shift the genus to cover species with elaborate labella, don't you think?

Hopper & Brown [25, p180] consider that Mark Clements actually lectotypified Caladenia carnea [25, p21] not Pfitzer and his C. flava. So, they reinstated Caladenia alata R.Br. Who is right? The taxonomists may eventually come to consensus after close analysis of the International Committee on Botanical Nomenclature rules but meanwhile the Column for one, is more or less happy reverting back to Caladenia for plants in the same morphological genus as C. carnea and C. catenata, i.e. 2 rows of disc calli not extending right down the midlobe, including C. alata.

But we haven't finished. Those ten classifications underlined above may not be the last word. The Tasmanian distribution of C. alata s.s. [26] stretches down to latitude 43° 25' on Bruny Island. That is as far south as Kaiapoi in NZ! We all thought Rainbow Mountain (latitude 38° 19') was a long way south for C. alata, didn't we? That big difference in latitude and climate says that NZ and Aussie C. alata either have to be different taxa or else ours hasn't had enough millennia, since blowing in to Northland, to move further south than Rainbow Mountain. What do you think?

Perhaps molecular studies comparing Aussie's and NZ's C. alata would clarify this matter? Even though it looks the same as the Tasmanian species, it may well be correct to call it Caladenia exigua Cheesem. The Column is plugging for C. alata R.Br. in NZ. Does anyone want to argue?