The Kranz syndrome in the Eragrostideae ( Chloridoideae , Poaceae ) as indicated by carbon isotopic ratios *

1!C/'-'C ratios are generally regarded as being very reliable indicators o f Q or C4 photosynthesis. These relative carbon isotope ratios are expressed as a negative 6 nC and fall into two distinct groups: Kranz (or C4) plants with Ó between -9°/no and -18°/«u and non-Kranz (C ,) plants with 6 between -22°/TM and -28"/no. In this paper, 29 taxa, representing 12 genera, of the tribe Eragrostideae were examined by mass spectrometry for their é'-'C in dried leaf tissue. Ail these taxa proved to be C, plants with 5 ISC values ranging between -13,6°/oo and -10.9°/oo. These findings confirmed published leaf anatomical observations which showed that all the studied taxa had characteristic Kranz leaf anatomy.


INTRODUCTION
A syndrom e o f anatom ical, cytological and phy siological characters, all related to aspects o f the car bon fixation process, has been reported from several families o f A ngiosperm s, This syndrom e has been called the K ranz syndrom e, C 4 photosynthesis, and the Hatch-Slack pathway.Kranz plants exhibit a high degree o f efficiency in the utilization o f am bient C 0 2, resulting in m aximum photosynthetic produc tion at high tem perature and high light intensity.T he anatom ical characteristics o f this syndrom e have b een known for 100 years (H aberlandt, 1882) but the physiological aspects have only been known for nearly 20 years (K ortschak et a!., 1965 andH atch &Slack, 1966).
T h e anatom ical specializations o f the K ranz syn drom e include the presence o f a chlorenchym atous bundle sheath o f large, thick-walled cells containing specialized chloroplasts distinct from those o f the mesophyll -they are a greater size, a greater num b er p er K ranz cell are present and they accum ulate starch-In addition, the mesophyll cells adjoin the bundle sheath and are radially arranged, Plants pos sessing the Kranz syndrom e fix carbon initially into four-carbon acids (oxalo-acetic acid, malic acid and aspartic acid) in these mesophyll cells (H atch & Slack. 1970).
T he only way to prove definitely that a plant is K ranz o r non-K ranz is to investigate both its physiol ogy, eith er directly, o r indirectly by m eans o f 13C /12C ratios, and to exam ine its anatom y, particularly the leaf anatom y.Differences in carbon isotope ratios betw een Kranz and non-K ranz plants are well docu m ented (Fritz & F ontes, 1980).6 I3C values fo r C 4 plants range betw een -18°/oo and -9°/oo with an average o f about -12%o, whereas the 6 I3C fo r C 3 plants is betw een -38°/oo and -22°/oo with an t Museo Argentino de Ciencias Naturales.'Bernardino Rivadavia', Buenos A ires.Argentina.
average o f -25°/oo (Fig. 1).In the P oaceae, no ra tios betw een -18°/oo and -22°/oo have been re ported.Plants with Crassulacean Acid M etabolism also have high carbon isotope ratios (Fig. 1), but are not relevant in this study since C A M has not been dem onstrated in the Poaceae (B row n, 1977).
T he determ ination o f carbon isotope ratios is a very' convenient m ethod fo r classifying plants as be ing Kranz o r non-Kranz.Only small am ounts of plant tissue are required, the d ata are very accurate and objective and the plant m aterial need not be physiologically active.T he use o f dried specim ens from herbarium sheets is, therefore, possible.T he genera o f the Eragrostideae exam ined, a r ranged alphabetically, are given in T able 1.All voucher specimen sheets from which m aterial was taken have been annotated with labels stating the 13C/J-C ratio o f that specimen.

METHODS
A bout 8 mg o f the leaf sam ples w ere dried under vacuum at a tem perature o f llO^C fo r 8 hours o r overnight.Every sample was then m ixed with 100 mg vanadium pentoxide (V 20 5) in a q u artz vial and flame sealed under vacuum of about 10-4 m bar.The vial, containing the sample and the vanadium pen toxide (V 205) was allowed to react at 1000°C in an jn_ J E
-1 0 T he precision o f m easurem ent is ± 0 ,l ()/oo of the ratio.If 6 IJC is > 0 the sam ple is heavier than the standard, if less than 0 the sam ple is lighter, o r depleted, in com parison with the standard.From this definition 6 IJC PD B = 0.
C arbon occurs in nature as two stable isotopes '-C and 13C , with the following average abundance: 12C : 98,89% !3C : 1,11% H ow ever, the l3C /|2C ratio is not constant and undergoes variations o f a few parts p er thousand in this ratio due to som e physical and chem ical p ro cesses.This variation in the isotopic com position is known as 'isotopic fractionation'.O ne o f the most im portant processes in nature which causes isotopic fractionation o f carbon is photosynthetic carbon as sim ilation by green plants.

RESULTS
T he Ó i3C values o f the taxa exam ined in this study are presented in T able 1.These carbon isotope ra tios are also com pared with published anatom ical observations.As always, in the Poaceae, the 6 13C values corroborate the leaf anatom y and all taxa have C4 carbon isotope ratios (betw een -I3,6°/oo and -10,9%o) which agree with published ac counts o f Kranz leaf, and stem , anatom y for these taxa.This study has, therefore, confirm ed the pres ence o f the Kranz syndrom e in these selected re p resentatives o f the E ragrostideae.--------------------------   Specim ens are housed in the N ational H erbarium , P retoria (P R E ).C hrom osom e counts w ere m ade from m eiotic squashes in aceto-carm ine (D arlington & LaC our, 1976) T he different Rubus species varied in regard to their geographical distribution.

a) T he subgenus Eubatus
The m ost w idespread o f the introduced Moriferi was Rubus affinis.Specim ens representing R. affinis w ere collected in the n orthern T ransvaal, Swaziland, N atal and southern and w estern C ape (Fig. I).H ow ever, the w estern C ape is the only place w here this species invaded the natural vegetation and where this species is regarded as a w eed.T he som atic chro m osome num ber o f 28 for R. affinis (G ustafsson, 1933, 1939 & 1943; H eslop-H arrison, 1953)   R. pascuus was collected in the eastern Transvaal (Fig. 1).This species frequently hybridized with R. longepedicellatus and the hybrids are included in R x proteus.Since R. pascuus contains both triploid and tetraploid specim ens, it may either represent a hybrid rath er than a parental form o r a diploid form might be present in South Africa.

MATERIALS
Leaf tissue was taken from herbarium sheets.This plant m aterial was obtained from m any sources, and these are all gratefully acknow ledged.Sources in cluded: D r Thom as R-Soderstrom , US National H erbarium .Smithsonian Institution, W ashington, DC (U S); M useo A rgentino de Ciencias N aturales 'B ernardino Rivadavia' (B A ); H erbario G aspar X uarez, C átedra de B otánica, Facultad d e Agronomia, Universidad de B uenos A ires (B A A ); Instituto de B otánica Darwinion (S I); Instituto M iguel Lillo, Tucum án (L IL ); Instituto d e B otánica Agricola, IN T A , C astelar (B A B ); Instituto de B otánica, Fac ultad de A gronom ia, U niversidad Nacional del Noreste (CTES).
ACKNOW LEDGEMENTS The authors are indebted to D r E nrique Linares for facilities m ade available at the IN G E IS and to the staff o f the Stable Isotopes L aboratory.Miss C ristina M aetakeda (C O N IC E T ) is thanked for technical assistance and D r R. P. Ellis (Botanical Research Institute, P retoria) for reading the m anu script.This research was supported with C O N IC E T funds.
The genus Rubus in South Africa.I. Chrom osom e numbers and geographical distribution of species J. J .SPIES* and H . D U PLESSIS* Keywords: chromosome num bers, geographical distribution, hybridization, polyploidy, RubusABSTRACTThe geographical distribution o f 14 o f the Rubus species in South Africa is presented.Chromosome numbers of nine o f the species were determined: six for the first tim e, o n e is confirmed and additional polyploid levels are described for the o th er two species.It is dem onstrated that the South African species o f the subgenus Idaeobatus contain less diploid specimens and more polyploid specimens than their extra-African counterparts.This phenomenon could be attributed to hybridi zation betw een the subgenera Eubatus and Idaeobatus.IN TRO DU CTIONT h e genus Rubus is spread o ver all continents and is found in m ost clim atic regions.Focke (1910Focke ( -1914)  )   divided this genus into 12 subgenera o f w hich only tw o a re present in S outh A frica.All the re p resen t atives o f the subgenus Eubatus Focke are intro d u ced , w hereas the subgenus Idaeobatus Focke con tain s b o th indigenous and introduced species.Introduced Eubatus species all represent the sub series Suberecti o f the series Moriferi and include the species Rubus afftnis W h. & N ., R. cuneifolius P ursh ., R. pascuus Bailey and R. flagellaris Willd.T he introduced Idaeobatus species.R. niveus T hu n b .and R. phoenicolasius M axim ., form part of th e series Nivei o f the section Idaeanthi.T he in digenous Idaeobatus species include the sections Ro st folii (R. rosifolius S m .),Afromontani (R. immixtus C .E .G u st.) and Idaeanthi [various species o f th e se ries Afroidaei, including R. apetalus P oir., R. intercurrens C .E .G u st., R. longepedicellatus (C -E .G u st.) C .H .S tirto n , R. pinnatus W illd., R. rigidus Sm ., R transvaliensis C ,E .G ust, and R. ludwigii Eckl.& Z ey h .. with its subspecies ludwigii and spatiosusC .H .Stirton].F o r th e purpose o f this study R. adolfl-friederici E ngl., R ecfclonii Focke and R. exsuccus Steud.a re included in R. apetalus Poir.b e cause integration betw een these taxa renders the separation in to distinct species impossible.Plants collected from a hybrid swarm in eastern Transvaal are included in R. x proteus C .H .Stirton.T h e problem s with Rubus taxonom y in South A frica are aggravated by the occurrence o f apom ixis, hybridization am ong indigenous species and betw een indigenous and introduced species, the va riation produced by a breeding program and subse q u en t escape from cultivation o f those plants and in ad eq u ate collected herbarium m aterial.T he aim o f this study is, th erefore, to provide cytogenetical evi dence for th e species delim itation in th e South A fri can species o f Rubus.T o achieve this goal, the re sults o f a prelim inary study o n chrom osom e num bers and species distribution o f the m ost im portant species a re presented in this paper.O th e r papers in this series will include studies o n m eiotic chrom o som e behaviour, reproduction, hybridization and will be concluded with a cytotaxonom ic study o f the genus Rubus in South Africa.MATERIALS AND M ETHODS T h e plants used in this study w ere collected throughout South Africa and transplanted under quarantine in th e Pretoria N ational B otanical G a r den.T he following 35 plants, representing nine dif ferent species, w ere used: Eubatus Rubus affinis: TRANSVAAL.-2329 (Pietersburg): Dap Naude Dam (-D D ), Stirton 5746.R. cuneifolius: NATAL.-2929 (Underberg): 14 km from Swartberg to Underberg (-C D ), Stirton 8154.2930 (Pieterm aritzburg): 3 km from M idmar Dam to Lions River (-C B ), Henderson & Gaum 93; 5 km from Pietermaritzburg to Mooi Rjver (-C B ), Liengme s.n.3029 (Kokstad): 11 km from Harding to Weza (-D B ), Stirton 8102.R. pascuus: TRANSVAAL.-2430 (Pilgrim's Rest): 1 km from Graskop to Sabie (-D D ).Henderson & Gaum 18, Stirton 9800.9861 & 9868.R. flagellaris: TRANSVAAL.-2530 (Lydenburg): Kaapse Hoop (-D B ), Henderson & Gaum 2. Idaeobatus R. apetalus: TRANSVAAL.-2430 (Lydenburg): Kaapse Hoop (-D B ), Henderson & Gaum 6.

R
. cuneifolius is restricted to N atal, with the m a jority o f specim ens collected in w estern N atal (Fig. I).In addition to the chrom osom e num ber o f 2n = 14 for R. cuneifolius reported by S hoem aker & Sturrock (1959), polyploid forms with 21 and 28 som atic chrom osom es w ere observed during this study.
Two R. flagellaris specim ens w ere collected in the eastern Transvaal (Fig. 1), This species has a somatic chrom osom e num ber o f 28 in contrast to the pub lished chrom osom e num bers o f 56 (F aasen & N a d ea u , 1976) and 63 (E inset.1947).T he geographical distribution o f R. affinis and R. cuneifolius (Fig. 1) indicates th at these species oc cupy different habitats.This suggests that they were differently adapted to the South African clim ate.T he single specim ens o f R. affinis collected in T ran s vaal, Swaziland and Natal might suggest separate in troductions.T he occurrence o f polyploidy in the subgenus Eubatus is restricted to eastern Transvaal and Natal.b) T he subgenus Idaeobatus The introduced species o f the subgenus Idaeobatus have a limited distribution.R. niveus is restricted to Swaziland and the adjacent Transvaal areas (Fig.

TABLE 1 .
-Chromosome numbers o f some South African Rubus spccies Rubus species (Table1).Poly ploidy occurred in six o f the nine species studied.