The genus Dipogon ( Leguminosae — Papilionoideae )

A taxonomic revision of the genus Dipogon Liebm. is presented. Only one variable species, D. lignosus (L.) Verde., is recognized.

A monotypic genus found in the Cape Province from the Cape Peninsula to Grahamstown.

Dolichos gibbosus
C liff.360.There is no specimen o f D. lignosus in the Linnaean herbarium, so one is forced to look else where for a type.The first possibility is the cited plate in the Hortus Cliffortianus.This cannot be regarded as a lectotype, hoewever, because the figured plant lacks fruit and does not have capitate flowers.L in naeus actually stated in the text that his specimen did not produce fruit ('Absoluta florescentia absque fructu periit').There is, however, a specimen in the Hort.C liff which one might consider as a lectotype, since it is quite sterile.But it shows no sign o f having borne an inflorescence and consists mostly o f newly flushed growth; neither does it match the figured plate accompanying the Hort.C liff, description of D. lignosus.The Hort.C liff, specimen could be a Dipogon, but it is very difficult to be sure.For these reasons it is an unsuitable choice for a lectotype.The remaining possibility is Linnaeus's reference to Eichrodt's Hortus Carolsruhensis.However, he does not mention the fruit and, as the phrase name is vague, it also does not satisfactorily resolve the problem.
There are two noticeable changes in Linnaeus's Sp.PI. and Hort.C liff, accounts.In Hort.C liff, he states 'Crescit in America', whereas in the Sp.PI. he omits the origin of the plant altogether.This may in dicate that his view' of the species had changed, as it certainly did in the second edition of the Sp.PI., where he included in synonomy 'cacara.s. phaseolus perennis' (a form o f Lablab purpureus) and stated for the species as a whole ' Habitat in India' .Verdcourt (1971)  Voluble perennial arising from underground, verti cal, deeply lenticelled rootstock with dichotomously branching laterals.Shoots spirally twisted, up to 3 m long, weak, glabrescent.Leaves trifoliolate; stipules up to 6 mm long, basifixed, oblong-lanceolate, per sistent, clasping but patent when old; petioles up to 5 cm long; leaflets 2-7 cm long, 1-4,5 cm wide, late rals smaller, stipellate, eglandular, paler beneath, glabrescent; terminal leaflet ovate-acuminate, late rals gibbous on lower margin.Petiolules up to 3 mm long.Rhachis 1-2 cm long, puberulent, armed with two persistent acrorhachial stipels.Racemes 5-10 (-33)-flowered, axillary, up to 25 cm long, longer than leaves, shortly and densely racemose towards apex, sometimes twisted; pedicels 5-10 mm long, somewhat flattened, becoming purplish with age, armed with small caducous bracteoles near apex; bracts small, green, persistent until flower abscises.
D. lignosus is endemic to the Cape Province and extends from the Cape Peninsula as far east as Grahamstown (Fig. 2).Its overall distribution falls with in the fynbos.Flowering begins in July, reaches a peak in October then declines rapidly after Decem ber.This species grows commonly in scrub forests, along the perimeter of high forests (Galpin 3988 reports one plant overtopping a 6 m tree), and more recently it has been reported from waste places and gardens.
Despite a number of recent investigations into generic affinities in the Phaseoleae, there is still no agreement about the affinity of Dipogon to other genera.Lackey (1977a) placed Dipogon in the sub tribe Phaseolinae Benth.between Alistilus N. E. Br. and Dolichos L., having suggested earlier (1977b) that Dipogon was closely related to Lablab Adans.and Alistilus and should perhaps be united with them.Baudet (1978), in contrast, placed Dipogon in the subtribe Phaseolinae, section Phaseolastrae, but Alistilus and Dolichos in section Dolichastrae.Marechal, Mascherpa & Stainier (1978) suggested links with Lablab.They included both Dipogon and Lab lab in their numerical analysis of the Phaseolus -Vigna complex.The peripheral affinity o f Dipogon and Lablab to Phaseolus and Vigna (Marechal et al., 1978) would rather suggest, as already alluded to by Lackey (1977a) andVerdcourt (1970), that Dipogon, Alistilus and Lablab may be better considered allies o f Dolichos.However, until the African represen tative o f Dolichos are better know n, this problem will have to be deferred.
The following herbarium material is recognized as D. lignosus:
D. capensis sensu Thunb., Prodr.FI.Cap.: 130 (1800), non L.,•Botanical Research Institute, Department of Agriculture and Fisheries, Private Bag X101, Pretoria, 0001.plant has become quite widely cultivated and has even begun to assume weedy habits in Australia, that its correct status be established.Freeman(1918)andVerdcourt (1971)have given exhaustive accounts of the origin, usage and misusage of this name.I will therefore not repeat their arguments here, but briefly outline why I have chosen t.21 in Sm ith's Spicileg.Bot.as the neotype even though the key diagnostic character in Sp.PI.726 (1753) 'leguminibus strictis linearibus' is not depicted in t. 21.The protologue in Sp.PI.726 (1753) is as follows: 'lignosus.9. Dolichos caule perenni, p e d u n c u lis c a p it a t i s , has described the subsequent confusion which thence accompanied the names Dolichos lab lab and D. lignosus.The second difference in the Hort.C liff and Sp.PI. accounts is the inclusion in the latter o f a description o f the fruits.The whole problem o f typification of this species hinges, I believe, on Linnaeus's uncertainty about the nature o f fruits in Phaseolus and Dolichos (Vigna was not known at the time as such and was treated by him under D. lablab and D. lignosus; see note by L in naeus on p. 1015, Sp.PI. 2.).It began when L in naeus's concept o f his species D. lignosus changed between his Hort.C liff, treatment and his 1st edition of the Sp.PI.I am fairly certain that when he added 'leguminibus strictis linearibus' he must have seen a fruit o f a Vigna and not one of Dolichos lignosus.It