Miscellaneous notes on the genus Pelargonium

It is pointed out that the name sect.  Pelargium DC. must be replaced by sect.  Pelargonium , and Pelargonium cardiophyllum Harv. (1860) by P. setulosum Turcz. (1858); P. mossambicense Engl, is reported as a new record for the Transvaal; a new name P. ternifolium Vorster is given to  P. trifolia turn Harv., nom. illeg.; and the typification of  P. multificlum Harv.,  P. plurisectum Salter,  P. dolomiticum Knuth and  P. dasyphyllum E. Mey. ex Knuth is discussed.


NEW NA M E FOR PELARG ONIUM TRIFOLIATUM H ARV.
Pelargonium ternifolium Vorster, nom, nov.Harvey (I.e.) also described a variety laciniatum, having the 'leaves 3-lobed, the lobes laciniate and toothed'.However, both undivided and deeply lobed leaves may occur on the same plant and, with our present knowledge of the species, it does not appear justified to recognize the variety as a separate entity.

T Y P IF IC A T IO N OF P E LARG O N IU M M U LTIFIDUM H A R V . A N D P. PLU RISEC TU M SALTER
With the original description of Pelargonium mullifidum Harv. in FI.Cap.1: 282 (1860) two speci mens were cited, viz.Drege 9460 and Zeyher 2054 ('Herb.Sond.').From this citation the impression is gained that two syntypes are involved.However, both numbers appear on single sheets in S, G and P, on two sheets in G and P respectively on the same label, whereas only the number Zeyher 2054 appears on sheets in P and W. The probable solution is provi ded by one of the three sheets in P, which bears both numbers as well as the inscription 'Legit Zeyher.Communicavit Drege .
It therefore seems as if only one collection is involved, namely Zeyher 2054 which was distributed under the number Drege 9460.
The material of this species in Harvey's collection (TCD) consists of three small branches of Zeyher 2054 and a tiny fragment marked Drege 9460, but without any indication of locality of origin so that it is unlikely that this sheet was used solely or mainly for compiling the original description.Sonder's main collection, referred to by Harvey, was acquired by Stockholm in 1875 and there is a sheet in S containing both numbers as well as the locality and name written on in Sonder's handwriting.We therefore designate this sheet as lectotype of Pelargonium multifidum Harv.( 1860), non Salisb.( 1796), as well as of P. plurisectum Salter (1941) which replaces the former name.

N E O T Y PE FOR PELARG O N IU M D O LO M IT IC U M K N U T H
With the original description of P. dolomiticum Knuth in Bot. Jb. 40: 71 (1907) it was stated that the type (Engler 2889) was housed in B ('Typus in herb.Berol.!').This specimen can no longer be traced in B, and it is presumed to have been destroyed during World War II.Neither could any duplicates be traced.
Although herbarium material of this species was borrowed from all the major herbaria in South Africa and Europe, it transpired that these collec tions do not contain any specimens with duplicates suitable for typification purposes.It appears as if the petals are easily shed during the preparation process, so that complete specimens with ample flowers prov ed to be very scarce.We finally decided to designate the following specimen as neotype : Orange Free State, (2926): Bloemfontein (-AA), J. W. M ostert 1661 (PRE!).This is not a very neatly prepared specimen, but it has more flowers than most seen by us, and it is com plete with mature fruits and part of the perennial basal part of the stem.It conforms with Knuth's original description, it is representative of the species and provides a good impression of the species in general.
Even though no type material survives, there is no doubt about the identity of P. dolomiticum.The original description did not mention that there are four petals only, but it did mention that the posterior petals are up to 20 mm long.This serves to distin guish P. dolomiticum from P. senecioides L'Herit.which has five petals not longer than ca. 9 mm, (with which P. dolomiticum is otherwise likely to be con fused).In fact, even Knuth in Pflanzenr. 4, 129: 401-402 (1912)

TH E D E L IM IT A T IO N A N D T Y P E C O L L E C T IO N O F PELAR. GO NIU M D A S Y P H Y L L U M E. M EY. EX KNUTH
The name Pelargonium dasyphyllum first appeared as a nomen nudum in Drege fs Zwei pflanzengeographische Dokumente, pp. 60, 90, 91 and 209 (1843).Harvey, in FI.Cap.1: 279 (1860) listed it in the synonymy of P. crithmifolium J. E. Sm.Knuth, in Pflanzenr. 4, 129: 373 (1912) acknowledged it as a separate entity and validated the name by providing a description and indicating Drege 3245 as a type.In addition to the type, Knuth cited several other collec tions.
Of the material cited by Knuth, we have only been able to trace Rehmann 2776 and 2823 (both in Z) and two type sheets.The Rehmann specimens are very poor fragments impossible to identify positively, and were collected in the Hex River Valley, which is far outside the distribution area of P. dasyphyllum as in dicated by other collections.Of the type sheets, the first, in P, consists of a fragment which appears to be the upper portion of a branch which branched rather profusely, with a dense mass of leaves and a single unbranched peduncle with a three-flowered pseudo umbel.The second sheet, in PRE, consists of a single short, branched inflorescence branch, and came originally from P. It is not evident that it is a frag ment of the specimen in P, so that it must be con sidered to be a duplicate.While these two fragments are hardly sufficent to provide an image of the species, Knuth's description could have been com piled from them, except for the fruit which is not evi dent on either of the sheets.
During our investigations of live material, both in the field and in cultivation, we separated a number of specimens of what we consider to be a separate spe cies from P. alternans Wendl.These specimens seem to match the description and type of P. dasyphyllum.
According to our observations, P. dasyphyllum at first glance is morphologically very similar to P. alternans and only remotely resembles P. crithmi folium.It is, however, readily distinguishable from P. alternans, to such an extent that we have no hesi tation in recognizing it as a separate species, as set out in Table 1.The branched inflorescence branches are nevertheless reminiscent of P. crithmifolium, and probably signify a close relationship to that species.
The length of the pedicel and the hairiness of the leaves are characters which usually hold good, but these cannot be used exclusively to distinguish the two species.P. alternans occasionally resembles P. dasyphyllum in respect of these characters and it is recommended that a combination of characters be used for this purpose.
The type locality of P. dasyphyllum was stated by Knuth (I.e.) to be 'near the Copper Mountains'.However, neither of the two sheets of the type collec tion which we have seen, bears that inscription.The sheet in P is labelled 'Camdeboosberg' on a label dif ferent from that bearing the number.This is pro bably wrong, as Camdeboosberg is far outside the Attention is drawn to a sheet in S which has the words 'Zwischen Zilverfontein, Kooperbergen and Kaus' typed on the label, and the name 'Pelargon.dasyphyllum' written on in Meyer's handwriting.This may well be a duplicate of the type collection, but it lacks the number cited by Knuth.The specimen is a small fragment of poor quality.

'
Department o f Botany, University o f Stellenbosch, Stellenbosch, 7600.**Hanks & Sm all, N .Am er FI. 25: 23 (1907) also designated a lectotype species for the genus Pelargonium, but for reasons ex plained elsewhere (in press) this is not acceptable.

A
C K N O W LE D G EM E N TSW'e are indebted to the following instances and persons : the C.S.l.R. and the Research Fund of the University of Stellenbosch for financial support; Mr C. H. Stirton for determining for us the precise date of publication (7 September 1858) of the name P. setulosum, while stationed at Kew as liaison officer of the Botanical Research Institute; Mr P. Drijfhout, horticulturalist in charge of our live collection of Pelargonium, who collected information on the sec tion Otidia of Pelargonium both in the field and in the greenhouse over several years, and who drew our attention to a number of characters separating P. dasyphyllum from P. alternans.U/TTREKSEL Daar word op gewys dat die naam seksie Pelargium DC. vervang moet word deur seksie Pelar gonium, en Pelargonium cardiophyllum Harv.(1860) deur P. setulosum Turez.(1858); P. mossambicense Engl, word gemeld as 'n nuwe rekord vir die Trans vaal; 'n nuwe naam P. ternifolium Vorster word gegee aan P. trifoliatum Harv., nom.illeg.; en die tipifikasie van P. multifidum Harv., P. plurisectum Salter, P. dolomiticum Knuth en P. dasyphyllum E. Mey.ex Knuth word bespreek.