Geology of the South of Magadi

EmRepert No.ﬁﬂ61

GOVERNMENT OF KENYA

MINISTRY OF COMIMERCECOMMERCE AND INDUSTRY GEOLOGICAL SURVEY OF KENYA

GEOLOGY OFOFTI-IETHE AREA SOUTH OF MAGADI

DEGREE SHEET 58, N.W. QUARTER {with(with coloured geoiogicalgeological map) byM B. H. BAKER, B.Sc., F.G.S. Geologist

Eight Shillings —- 1963 u FOREWORD The publication of the report on the geology of the area south of Magadi completes the account of the southern end of the Rift Valley as it occurs in Kenya. The Magadi area itself was described by Mr. Baker in Report No. 42 (1958). During the mapping of the continuation of the Magadi area the discovery of some critical exposures enabled the correction of an error of succession in the lower Pleistocene rocks that had been made during the survey of ~ the Magadi area. The area is wild and desolate, but of considerable interest scenically, with the western Rift wall a little beyond its west boundary, rugged hills of ancient rocks in the south-east and two prominent volcanoes, Lenderut and Shombole, rising from the Rift floor. There is little mineralization in the area except kyanite in gneisses that outcrop prominently in the hills in the south-east. The content of kyanite does not, however, exceed 19 per cent in any of the samples tested, and in view of the low grade, inaccessibility and lack of water at the outcrops, the deposits are not commercially attractive.

Nairobi, WILLIAM PULFREY, 26th May, 1959. Ag. Commissioner (Mines and Geology).

----- I CONTENTS

PAGE[‘ -'\{'. [1. .-\'E‘-HL:".1L‘LAbstract

I-Introduction[Itirn‘ULLclH‘n . ._ _...... 1l II-Physiography-l"}1_\ §.i'\‘-§__':"i'."f‘."\ . . . 2“ III-SummarySL.i1‘.[]‘.LlI'_\' m'of (:L‘C-iOE.)Geology ...... Sr4 [ IV-DetailsLk‘luilh l‘i'of (icn‘-'::‘-g)-Geology.. ,. ._ _. ._ ._ ._ ...... 5 1.i. BasementBuscnwzn System-SVnEL'IH Tiny-kn.Turoka 53=_':'i:\'Series .. .. ._ _. .. 5 I 1]](I) Q's;Quartzites . . 6 J._II(2) KyaniteK‘glrnac gneissesGIL-1mm _. ._ ...... '57 (3):FI BiotiteBiotin: gneisses‘:'.:".._'é\-.i'h ureu'and biotite-garnet' gneisses 7

(4)4—?) MuscoviteNi 118.1.“ iLc gneissesLi'li'ét'n 8O’.‘ :5](5) Limestonel...i:T.u:c.-‘c 8

{m(6) GraphicGraphite schists4gl‘jsas 8’7“

2. Tertiary'I'cr2f and Quaternary Rocks.Rx"\.‘i\'§ ...... I“8 {11:(I) Volcanic\'u car 'Rm'ksRocks ...... _. _. .. .. S8 {m(a) Lenderut|._CI‘_L‘I-:ru[ Volcanics\ cluiraiL‘S .. ._ _. _. .. .. 5’8 m1(b) Swan”Shombole.1 VolcanicsVanuanics ...... k)9 (c)I;'.‘.- KirikitiKEY-":2: BasaltsBar .. .. 9 :rf](d) Oli‘Olivinefil‘c Basa}tsim ...... _. . 9 (e) Plateau Trachyte Series 10 (f) Ash and Scoriaceous lava of Alasho 11 (2) Lacustrine and Fluviatile Sediments 11 (a) ChertWSVI-cx'Series ...... 12I1 I132](b) H:L".{‘L"I‘.‘:ELlOloronga BedsBeck ...... 13i3 fr'](c) Pebblel’chbic Beds and Sands of Pagasi 14 {if}(d) Scum‘rcrlb‘.Sediments in fault-troughs 15 I'm(e) High Magadi.\- Beds and equivalent beds of the Natron Basin. . 15 [3|(3) SLLP-L'i'iicéul-Superficial DepositsDora->12» ...... _. .. . 1717 V-Structure..\ SLr'ilc: .. .. 18 VI-GeologicalVi Cicelogical History}-E'_:~'.l{.>1'}' ofU.” the MagadiRinggit-Si SL115-Section' of the Gregory Rift Valley 20 VII-Pleistocene\-"|I—---Plcistoccnc C'hmnn'nr}Chronology and Climate( _. 22 VIII-Mineral\'Il[--- Mincra? DepositsDepofifir . .. .. -- -- ...... 2333 1.].KyanKyanitefic ...... _...... 23 fl.2. Graphite sclzistsschists .. ._ .. ._ _. _. .. .. 25 ;. 3. Water—supplyWater-supply ._ . . . . _ . . . .. , . . . 25.25 4. EconomicF-Iconmnic possibilitiespmxibiiitics vi'of the arcs.area . _ _ ...... 26 , IX—IX-References.Rct'crcrucs _.. ._ ...... _. .. .. 262|" LISTLIH'I' OF()!-' ILLUSTRATIONSH.1.i'S'I'R.\'I'I()\H Fig. I.-Vertical-\'cr:‘.c£'.l accliimsection Bll'JL'J‘.\'i|"i.-_showing structure ‘n.-ibetweenMaven KakuKARL: 35:3.and 2_.i.'the Loldorolo{suldor'un j‘lliiihills (16 Fig. 2.- TltcThe paucrnpattern u.of faulting in the Magadi region 19 I..‘ .! fig. 3.-Geological(|:_.. l- _..-. sketch-map of the Losirua kyanite deposit 24

j I' ]

1‘ "'J MAPm. ,4. (.jev-gigalGeological MapMLm ofL‘l the Area.A 5Southx11?“ of Magadi;\Lngncﬂ: hicwr'cz;degree «Hm:sheet 58,TR north-west13 quarter (Scale\Smic 1:MFRK'ULHZM125,000) .H. ‘.. . H.. ...‘ ..‘ ‘.. ..H .. .. :11at cm}end

ABSTRACT 'ITheEtc reportrepvfir'. dcsm'describesfi an area 616 square miles in extent in the Kajiado District, south of Magadi.Mug: 'ii It[1 is boundedbrmmfixi by longitudes 36° 00' and 36° 30' E., and by latitude 2° 00' S. and the Kenya-TanganyikaKcnyn—Tgr‘gm;H11 in.boundary. The area covers the southernmost part of the Rift Valley in Kemxt‘Kenya. ‘\A grmmgroup mtof hilixhills (Hin 13‘;the \south-eastern‘11”1-Uném'ri ‘_"partJ'l .1of the area rise to above 6,000 feet. The central”911111 17'part'1 of 1116the 11mmarea is' 'broken C:by numerousI'MHHL‘I'UI' ~ “lb-“m1-sub-parallel youthful escarpments and is at Lu:an gelevationcwnon of 2_(2,000 to 4,000 feet. TheHr. dissected‘_. remains of two volcanoes rise to 3,500 and 55,000.000 feettbcr respectively.rc. Qumr‘witcs.Quartzites, 1biotite' ' gneisses and kyanite gneisses referred to the Basement System, outcropD in the hilly“.1133, south-eastern corner. Tertiary and Quaternary volcanic rocks com- prising trachytes,machylh. mmnephelinites, olivine basalts, andesites, tephrites and basanites, and lacustrine1C and fluviatileflux [3.11; .v'sediments occur in the remainder of the area. The volcanic, tectonic and sedimentary history of the whole of the Magadi section of the Rift Valley is briefly described.

SL5Sub-economicb—L‘con mm is but large deposits of kyanite'.‘C 1.11;are IJXL‘THmentioned“KLU d'n‘.and “Oi".some aspects of the dcwclnmucu:development N'of Mwater-supplies“Lc' are discussed.

I I ,I I I I I

'l it I GEOLOGYGEOLOG‘! OF THEEHE AREA SOUTH OF MAGADI I-INTRODUCTIONl—I'Vi RUDL ( HOV General.-The area south of Magadi as defined for this report is the part of the north- l west quarter of Degree Sheet 58 (Kenya: Sheet No. 171 of the Directorate of. ()1Overseascm Surveys) north of the Kenya-Tanganyika interterritorial boundary. It is bounded10.111 Lid-J by longitudes 36° 00' and 36° 30' E. and by latitude 2° 00' S. and the Kenya-Tanganyika boundary and is approximately 616 square miles in extent. Owing to the similarity between the present area and the Magadi areaLJCL. to:11 the111g north11012211 ,,~ (Baker, 1958*), the two reports should preferably be read together, especially 'I.as\ the‘ U111}oppor- tunity has been taken to summarize the geological history of the whole2111‘: c 1121I‘1:of the \1;1g:1111\cMagadi section of the Rift Valley in the present report. The area south of Magadi falls into the Kajiado administrative district of Southern Province, and is inhabited by nomadic and pastoral Masai. The only buildings are two African shops at Alangarua at the northern foot of Shombole mountain, and a third shop at the foot of Lesiri hill in the eastern part of the area. The only produce of the area is hides and skins, but a small ii.\21:'1_-.:fishing :::.2.'1\'..'>-.industry :~.is -.'-_'1‘1‘11'cdcentred on Alangarua, where catfish are caught and sun-dried on a small! 5211.11scale. 8.1.15Sales ”:22of bli‘stockﬁk organized by Government do not seem :11to '11.havey .' begun 2'1in 1211‘this '..1'(';.area ..1\as 11.;:1they 21.31:;have ¢21xc'-.I.|‘.:1'L'.elsewhere, probably on account of the poor L.2-!]112T.J|21Ecommunications.‘l Communications.- There are no roads in the area, only a few motorable tracks developed by intermittent use by traders and prospectors. The principal track leads southwards from Magadi to the south end of the lake and thence generally south-westwards.1: to Alangarua. A further track made by a prospector to 9.11gain1.:.1'.'s\1n.1access to a magnesite111..:-'1:.<.-1: ddeposit1.111521111111i1con the TanganyikaTa. 21111211111 side of the boundary leads from the south end of Lake1.112AL' \1:1g.1111Magadi 5111111‘1-“1111|1—\'I.1.>1I.\111115south-south-westwards across the area. A very rough and scarcely noticeable track leads16.11% fromfun; this11:. .magnesitemncx'll: deposit and enters Kenya due south of the Lenderutww .. hills and,. 2.11:1after I'L'.l{'2reachingH15 [21:the LesiriLL,» shop\lil‘l‘! in the east by a circuitous and rocky route, proceeds:'11-;¢_--.221\. 3011.11.5111.11%north-eastwards '..1'.21and u‘x-:1":1.1eventually joins12111.11: the Kajiado-Turoka-Mile 46 road north-east of the present area. A motorable route from Singaraini station (outside the north-east corner of the area) to Lake Kabongo is also shown on the map (at end). This old track is exceedingly rough and so little used that it is not visible over the greater part of its length. Some of the larger soil-filled valleys are motorable along part of their length, :1.1111L'1'}namely Rekereien, Sereata, Narsurana and the greater part of the Ngare Nyiro plain1:1 111.1x1u‘coutside [:‘1:the swampy areas. The lower Pagasi river and the southern end of the Kirikiti. 5.117111111'111escarpment c."can be reached by crossing the Ngare Nyiro by the steel bridge west of Magadi\i;1g..d. and 1111.1:driving. southwards along the‘ 121,102;foot 021of 1121Cthe 512‘Nguruman escarpment. In spite of the lack of maintained-2 112:2. 111.111)roads 1111and [1‘.-Cthe roughness of the few existing tracks it was found that no part of the area22L‘L. is.\ li‘ll‘il‘more ['121'1than .‘a :C'fewﬁ I”:l‘|.I2\hours walking from the nearest motorable route. Climate.-The greater part of the area is semi-arid in climate, 1.1.1111with 11-11111c1'.1:111'c\temperatures commonly exceeding 100° F. and moderate humidity, the latter rising to an' '. uncomfortableL;'1;‘-1111'.2111':'.11‘.‘11' level during and after rainstorms. Towards the south-east the ground rises steadily from 2,000 feet to over 6,000 feet among the Losirua and Kileu mountains. Here, and along the plain extending along the foot of the mountains the climate is pleasant with2:1111'1.:1'.11‘2-1\:‘;11more moderate temperatures and lower humidity. ,. Maps.-The only existing topographical map of the area available at the time of the survey was that of the Anglo-German Boundary Commission (Uhlig, 1909, pp. 9-10) whose map-Sheet 4, scale 1:loo,OOO-was found to be inaccurate. The writer2.1:." 1.11.1.2:made .1a 211M)hasty plane-table2L'5L2113I1'l11n'1:11111‘1C1101‘11‘17A7survey based on three major triangulation stations outside the area to the north- J Olorgesailie, }_':1En Doinyo12) Nyiro and 01 Doinyo Nyegi. The1:112 111.:151map 1'.L'u1113111.131).accompanying this report was then compiled using aerial photographs. Altitudes were measured:=.'.;.\L.E':.\-. by a single aneroid barometer and locally differed appreciably from the barometer LJ-CNCF‘.observations.iI 30H“; 11].:made by the Anglo-German Boundary Commission. The writer's altitude1211: 1101'}.work 13.21.».was 11115111based 021on LakeLukr Magadi, which has recently 3312.12been1 UL'LCT'determined1|1}\..l.|l L21“as 1,973 {131122feet 112111110above 11'1c'1111mean 511140.131.sea-level.

"References are quoted on p. 26.

... I~

The northernnn"1her:1 margin111.1: "'1 111'of the een'ngie.1‘geological map111.171 1.1:(at L‘iidliend) doe;does not:101 L‘L'1ci're.~11t11‘.L.correspond emet.exactly3 with the southern. nithei'n margin111111; of the '-_.'L"..1l5.1g1L.1lgeological map1111': of.11' thehe area.::':.1 t1.)to the north110t‘i\l=.1g.1dl(Magadi .-re:1,B'.-'LlLL‘1'.area, Baker, 1958)1951:) due to the fact1' t that the latterl'.1tL'5:.r 11:13was basedl7.l."i._'t_l 111in partr‘:2.rL 0:1on :i1ethe Anglo-German.5\11'glo—German Boundary Commission 111:nmap referredrel'eri'ei. to1.1 above,..l‘..1'. - whichhiL'h 1:11:the writerL "ite: found‘L'..'L'.11'...l toto behe inaccurate.i111ee11rt1te, The lackleek 01'of L‘or1'e5pondez13ewithcorrespondence with the map111.11‘ to..-. the15"1‘L‘ e351east 1V..L'11:1'1_,..—B.'s.

PreviousPrevinm ll'iii'.l..—Work.-SeveralSL'\L'1"..l L'.11‘l_\early e'\11lL'11'L'1'~.explorers passed,“L'.\\L"Ll along the basehttse 01of the Nguruman1\g111111111111 escarpmentescarpment, this 11.1.5.5being :1a well l~:1'LL1\'5.11known 10.111:route from11:71:11 the e:11'lie».11_;-_LL-earliest days L‘lof e\p-exploration.oration. The 11151first \‘was.1: FlseFischerher 111in NS}1883 1.11:."(see ('1Gregory:'e_'111'j. l‘é‘ll.1921, 1‘..p. 1791179) '.‘<..“5.Lland l1.11e."l".:1'..\e1'.later Kaiser, ‘11111but their obseobservationsﬂations were 1110lprobably'll\' 1151'made.1.LL" to the 111:."1l111.north of '.l1e;11'e\e1‘11'.11'e.1.the present area. ('.'.C. L‘hllgUhlig :11‘1L‘land F.1:..l1'1egerinJaeger in 1904l904 entered the area from the south 11ndand visited~151'L'L'Ll .Shombole'1L'11'11l'L'1le 111L11111.11:'1mountain and the PagasiPngtisi river '.'Lreziarea (tllg(Uhlig and Jaeger,.laeger. 1942,19.1: pp."1". 3‘131-43).4.31. TllL‘V'They LldescribedﬁhCl'll. LLl t‘1ethe s':L1'1;:1"..11h_\geography 1111'of the NgurumanNguttnhan escarpment and noted 1111'1Ld51'1n111earfvedsonite 11':L."1'1;.51etrachyte merlziinoverlain by yeila'xyellow 1‘..tuffs._s '.'L:1dand boulderLo Lllder hedsbeds :11at thethe southern end 01'of the eseurpmehtescarpment V'Ltsouth 01'of ‘.lthe.- Pagasii’:1e.5..\i 1'i\e.".river. \ieMention'. ti L111 was.1115 111:1Liemade 01'of the Ngare\gare Nyiro plain 101).(op. £11.,cit., 1111.pp. 3737-39)391 11ndand 1.1Lthe 55.11.11:swamp,11.1 .1:and11‘. 15.1they L.-:1.‘eLldescribed ‘.‘.‘ex'tL\:11'\—LlL‘.westward-dippingipping tollstuffs'andand 1.11-1.13lavas L111on Melil hill. 'l'l'ee}They :Llwalso 1.1.:ser'1'etobserved the beachbench 1:11:1eegterraces 0:1on Melil.Melil A.5\'1_illdeserip"11:1nfull description 01'of the 1L3pt.1g:'211‘h}topography L11'Shon1ht‘iieof Shornbole '.Lwas.1.\ 32?.givenen 12:11.(op. eit-cit., 311:.pp. --l41-43)[—1.1 1 .1:1Lland the 1".01'1he1'11northern face|':1ce oftheof the 11111111113111mountain reeogniyedrecognized :15as L'.‘L'.111‘~1‘L'1Sr:Llcomposed primarily‘:'11‘.1'_L1 ‘."l'of 111:"tuff :1121iand nephelinite l'.1\".1.lava. The rock collection made h\'by UhligL5 hll;._r :11'1Lland Jaegerl.1eue3‘ is' listed‘. 'in. .111an :11111L11L‘.1\appendix 15.111.(op. L'lt..cit., 111:.pp. l3!181-188).$881. The rocks collected from the 1111:111present :.area.Li‘. were1'e 11':'L‘l‘.).'tietrachytic t1::'1'tuff .1:11.iand '.1'.1L‘h_\'testrachytes from (N01 Doinyo Lengot‘ztle,Lengorale, thethe settthermnostsouthernmost hilleel.hillock 01'of the Nguruman1\Lv111'1'111:1.1 eLIe.'L1":111e11tescarpment immediatelyin1111edi:5.1L-'l_\ soothsouth zofof the PagasiI’ttgetsi river 1011.'L'11.,11.idliundrtephl1LLLLlt'..LV:Lt1Ll"1‘1~1_'51‘.1Sl1L1111l:.leLL_1p.c1(op. cit., p. 181) and nephelinite lavas and tuffs from Shombole (op. cit., p. 182).183.1. Capt. G. E. Smith 1191171(1907) passednamed through the :1:areaen '..1ie'and e.1r1'iedcarried out much sun-tyingsurveying en mute.route. He gLWCgave :1a gei‘iei‘dgeneral dL'SL'idescriptioni111'..'-:1 ol'of the NgurumanT‘Vgtii'11111.1v1 e5:escarpment1111111161111(pp.. 256—256-7).‘) :111andd the hot springs '.11at the 51111111south '.'1'.Llend 2.111-115Veof Lake Magadi.\l .LL11111(p. 258).W 1. Smith re1.11'ded1l'1e11m_1,rregarded the floor of the RiftRil't Valley 115as bl'O'ef]broken "11}"by 1:1:111e1'L111.\numerous Ll}lx'esdykes (_"lof 11.11‘Llhard lt'L'L't'.lava",.'L1lL1L'l1which hehetthought[1-l\ might gategave rise to Thethe hornt.horst 21ndand 11'L111ghtrough topogi‘ttph}topography.

P.P W.tWaltherhet‘ (‘.‘)t(1922) L‘L'fei'il‘LLldescribed :51ethe 1111111trona in lLake:LlL'e MegddiMagadi :.1‘.'Liand found 51151116saline incrustationsinet‘ustations some 35.25 111‘.mileslL‘s :Lmtlisouth 01of M-Magadi,L_'_".1L".5. 111'L'\1:11'.:L‘.‘l).presumably 11‘.in the :‘1'L‘11area 111'of the Ngareﬁgure Nyit'oNyiro S\\.:11T1p.swamp. They metewere reported as eonsisticigconsisting :11'of .~‘.sodium:Lli.11'.1 bicarbonate,l1iL"..rl.1:1'..1e 111:1.but \xhei‘when t‘1ethe .samplesmles were examined Zil'ICI‘after :1a eonsidewhconsiderablee51'11e1'1.L..linterval 'ltheyL". were. L15. LlL'..'11‘.1i:.:e'5determined '..sas 11'1'111:.1trona. \\'L'11‘L"1L.1'"L'1l51'3I'ICIJOI'LCdWalther also reported h-havingwing (lL‘LCC‘LCLldetected L‘L‘.1‘l.‘01‘1L.lD\lL".L'carbon dioxide '.__'.1.\gas L‘.\L‘.'LW111LT101111111111L1L11_.V'l1x5111‘e‘x111escaping from numerous fissures in Nullsoil :Lm‘land I‘OC‘KK.rocks.

II-PHYSIOGRAPHYlf—l’H‘) MUGRKPEIV ”IThehe '.'Lre:1area e111:can l.‘e11i'.1L.leL‘.be divided into1:11.. t5'.'1L'ethree physiographic111.1.511'5;.r.r"..1.‘11ie11'111'5.units, :he111.71.1111:=.inL11.:.~;the mountainous eo.11‘.11'_\country inin the south—eastsouth-east together '5.'L1Lhwith the gent]gently-sloping'1'——l111111;.' soil pedimentnediment to1.1 the nestwest L'11'i1'.of it, thethe Lent.central.1] 1.1mlava 21re:1area consisting of many111.111) :'1:11'1'Lnarrow'52 hhorsts.a'sts :L:1Lland 11'i.;1.gi1s.'.11.dtroughs, and the \gut‘eNgare {\311'11Nyiro pl'tjnplain together[age-.1191- with.1 th the V'Lititherninostsouthernmost '.1.and1.1 lowestloat .part'.. 1' . L‘l'of the Nguruman'.\":11. .'111 1.1111 Lse:11'1'~.111e111,escarpment. The 1111.11111111i110115mountainous south—23.1%south-east CDl'll‘L‘l'corner Lof1' 1‘the'1e Lites.area is Lion11r1'.LteLldominated h}by the KileuKileL: ridge and Kt‘tkttKaku 111L111111'.1i:1,mountain, ‘.'.hlehwhich teachreach '..lt:altitudestLL'Ll-:.~' {'11of 11nearlye: " "101115)7,000 1'eet.feet. TheI'he KileuKileti ridge is composed largellargely11 1:1of rL'gisL:resistant1111 L11.quartzites:11'. llZL‘S ‘\'l'll-'ll‘.‘_‘striking to 1.1L"the north-north-east,115:1 :1-1'.L11".h2L'-.1st. and .1a few1'e\\' milesmileﬁ north—“estnorth-west 1.11:are the l_o.\i1'Losirua11.11 .1ills.hills, 11..'11'emore or less-w rounded in onoutlineline .Li'LLland 1'e.1L‘hingreaching 1'.-earl);nearly to (3.0111)6,000 feetfeet. NorthNor'Lh 111'of Losirual.2.'.15i1'1;.-. .L;'L'-are s1smaller.111llL1 l'L'.‘-residuallL.LL.Ll L'quartzite'1.1.'1'51e hills"i5l5 51.611”.such as LesiriLL':\'11'3'111‘1Lland LCIlhljilL-‘C.Lenkijabe. Regarded 1nin '.‘1a broaderl.10:'111'e1' >e111ngsetting the KileuK1.e11 ridge.'1L.1: 'is ‘.l15-the >L~15.L‘.1southern1':'i L'L‘.1‘.7.111111'111011continuation 111'of thethe LL11111_jiLuanji hills \\l1l'L‘l1which extend 1".11‘far 15.).to the 1:1:1':l1-e;.551.north-east, .11111and the.'1..' LUSJI'LJL}Losirua Lynnegroup and its surrounding hills occur 11?":as "111.1tliers""outliers" 01'of the 1111111main l1i'l_\hilly ..re..Larea. 'i'heseThese hills hLLLL':have steep slopes and are at :-1a late)late-youthfulouthl'til stage 01'of Llissdissection.5.111. Tl‘They_ are' marginedred 1.1to the nestwest l..1\'by '.La gei1tl_v-Vlt_~pi11ggently-sloping 111:1plainin covered with grass and 1'1:'.1L'i1e.\patches L'l';;e.1ei.1of acacia forest.:'5..11'L'>.. This13113 plain,111:1'11‘. 5.11'11111L1of which M'tM'balbal:lhul Ormeresho is :1a part,part slopes 1101111down from 4,50114,500 to1.1 below 4.1'4,000".l lLLL.feet,"“ thell'z'- .soil" emercover beeon‘tinebecoming thinner xxes1\\'z.1rdswestwards, till 1321511115basalts 01"of the R111Rift \Valley11558:. emerge 11'.from1‘. henetttl:beneath it.11.

The central 1.part'.11'1 1'11'5of thehe .11'e.1.area, which' : '.‘is composed“ '1 setl L'11'11llL':Lliof alkali ti'.1el1_\'testrachytes and basalts.basalts, is closely gi'id—l'atiltedgrid-faulted along.ralong linesline-1V trendingtr'er'LL 15' approximately 1515. degrees emteast L'11'of north,11or1h, the resulting topography being 1hthat:11:of elongated ledges, platforms1113 '.111dand troughs1rL112gl15‘LwoL111de1.bounded by'by rockyoeky eseztrh-escarp- ments1111:1113 which are generally5.1e1‘1e1.1 'lf'yL''LLL.vertical‘ l in the11".“ upper111—111;.“11115'1.part. ErosionFrosioi‘. has made111::de littlelittlle impressionii11p1‘essiein on tl1esef1'11111—hloL‘lLsthese fault-blocks, but1111 lakel'. :lL'e beds11en 111'or Vswampy'..1111 ' 'soilxsoils frequentlyL'i'eqner'ii'ly oeenm'occupy the ﬂfloorsoors 01'of thethe troughs. 'Thehe Lenderutlenderut l‘Lihil1s,.LV.1\.l.L1sewhose summitsV1...11L1'11V 1rise 1,500.0 tLetfeet '.'.hL'1'.eabove the surrounding51111'1‘0.L::1Ll511;._I grid—grid-faultedfaulted countrvcountry, represent the remnants1"e111n:1:1t'< of1.11:1a denudeddL'LLLLLleL'lxvolcano,L'wl'L'.'.115.1. the lowerlove: partsnew of01' which11 11 eh haveh.22.10 been engulfed 3u)

‘0}by the :1ikztlialkali ts'zteitf-te'5trachytes '.:11Lland i1'...5".1lt.5'basalts tl'..1tthat CO‘1L"I'li'.Ccover the RiftRil't Valley\13ii‘3‘- floor.:lOL'ti‘. 'l'l".8'..\8r'.15_18The average levelletel 112-"of the 1.111grid-faulted—1'1111ltL"Ll 1118;:area 0858811115descends from1111111 4.0004,000 1881feet ini:1 tl'.8the 811.51east tot11 hL‘l'Lmbelow 1.0002,000 in the central and 111.8.5t'81'nwestern parts,part5. where\‘11i‘1L‘1'L‘ the lavas1111"»; pass1111.55 beneathi'VL‘HL‘LlI'H itilttiittmalluvium 111111and lakehike bedsl381‘15' belonging to Lakes .\'.1.1r111-1Natron 111111and Magadi.Kittgtttii. 'l'hi5This \=.L5t\\111'dwestward 5111118slope 55'is the result1'85'L1lt 81'of the:the tilting.1tilting 1'11'of many111111111 111'of the full“fault hioe'11'5blocks 111to the \'1.8.5't.west, '.1:11'land isi.5 :1l:'-1.1also 81118due to th."the 1118:1181greater 81.1111:1l;1ti\'8cumulative throw1111811 on west-facing1185141185115; faultl'attlit 858:11'1111181115.escarpments. - The Toroseiit'11'858i 1'21'81'river .tii'L‘iand itsit5 ti'il'tr.11.1..';.tributary the 51111.1Sinya LandariL.1:1Ll.1.'i 1'.L‘5.'L‘81".idescend from the western 11ndand northernnt'Jr'thei 1'1 t'L'1885'faces 0'1"of LosiruaL05ir'111't Mountain '.11'111and run '11westwards8511.1.‘11'81'5. .181'1'155across the grid-111113180grid-faulted topography to end in111 the N111‘:1'~L.'.'t1'1'.tNarsurana '1:1ll8_\'.valley, in '.hethe l811185tlowest part11.112 L1:"1'1i'1iehof which there isi5' '11a 581151111111seasonal 511311111.swamp. The riverti'\8t tun-81585traverses the TLtitJSCiTorosei, Kisielibl\'15i8lil. 1:110and N.11‘5L::'.11'.'..Narsurana valleys,‘111iiL‘_\'5'. '1'xhiL'hwhich i'.it has partly11'.tr‘tl_5' infilledi:1lil181111'ithwith r! . 5' 118711 111.5. 1111.1 ‘11:'8: 1l1'5' Ii‘lt'L‘LlLIil '.hL- 111181'1811i1':_l.11.1 .) its detritus, and breaks through the intervening lava horsts111111.515 in111 narrow11.1.1'1'11-11 gorges131811585 uhiehwhich are:'11'8 as muchLl11.15as 200-‘OOlCfeet8t1J8L'11.deep. ()thL-t'1'.\Other rivers81'5 Ll8.5L8nLiingdescending from”11111 the high ritlgc5ridges 81151east 111'l.85'i1'i'11111lof Lesiri and Lenki-l..8nl1'i— jabe_i'Lt'DCi hiilrhills traverset-"ttt-L‘r5'8 the plain1ti'.>.i11 butl111‘. 8111iend in it. Thelite 'IL::'11‘11.1Turoka river,1'i ' . however,htJ'1'1-L'wer. 81'1t8t'5enters theti18111'e::area in the north-eastnorth - " corner' and 1111.1."1following11.8 the1118 5812:south-south-westerlyh—5111.1111—185181'l} mendingtrending fault1"1'111lt t1't1t1gh5.troughs, 81105ends in Lake Kabongo,I5'.:1i.'11_111g11.11whichhiLh‘ occupies1>L81' {‘11135' ..a fault-trough.1Lli't— 11'111111‘11. A-\.1.11'tl181'further,. unnamed,111111111181.l river1'i\'81'1l1';ti1'.5thedrains the CLIStCt‘tleastern slopes of the Lenderut hills .11'.'.l.'l.Land the h10.11.ibroad dip_1 .5l1'118slope otof lava1.11.11 8.11.51east 1"l'0'tl111n11t.of Oldomut. Itlt 5."Ki1'15skirts the eastern Lenderut‘ -.11181'111 lL1Lttl'1il'l5'foothills '..111l8and endsml: in511 thefh. UiLiOlllLllOldomut 11?.1i11.plain, the.the 5118site '11]of a1; former1‘111'11181' lake,l'.tl1'8. 111113.11which overflowed8.1 t."to the13181north-westMirth—11851 intoi:1t11 the'.he .5111".l'1-8'.1.5'181'11south-eastern lagoon‘-l1¥1l'-‘l7 1\l'of Lakel '.158 Magadi.\lttgt'tdt.

5 r. The principal fault-troughsl'.1'.1l1—tt'L1tt1_L’l1.5 '.118are l5'.'_'Kisielib, theF: long,'l:_'. 11:11'1'1‘11‘.‘narrow .111Liand LiCL'Ddeep Ser8.1'.'.~.—l1;iatml)t1Sereata-Kiambu \".t.‘|Lj.'.valley, .\':15‘.tt"Nasurana,111. the i.1_‘5'ttii-R‘._'.‘i\Cl'CiCllLesoit-Rekereien .11.trough. which is1.5 ti.the 5.1'111'181'11southern 8111'12111131-1tioncontinuation of thethe LakeLuke MagadiMztgatii 11'L111trough,rl. 111111and the L'1z'1.1.1dbroad Ngare.\1 1\'_1i1'11Nyiro 0813185511.:111.depression. l-x'il111:1gz1i1'.i.Kibangaini, :1a largelarge flflat-at— floored:lL1L18'L'ld1'i81l'tl-1'8L8Ll.iSl-.1r§__’8l;1dried lake bed, is largely ini11T11h1::.1Tanganyika-only_ 15.1 1111531 its115 northern 11.11tip 81881115occurs 1in H18the present 1".lL't-i.area. The T5g.11'8Ngare Nyiro.\'_1'i1'111.1l'.111'.:1:Ll.5'1.1.1:1.:;11118.11‘81.plain and swamp are a few tens.8115 L'1l'l'88111l11118the11185811138181811'L.1l18of feet above the present level of Lake Natroni\'.'1‘.rtin 11118..and, in their 51111‘1‘1-81'11southern 11.1:tpart '1'185"west '.11'11‘.and north-west111.11'1‘11—11L5t '01of ShShombole,unihtle. tiiethe plain1.111.111 is5 CL'tlTlDOSCL.composed Llof lake11111'8'r181'bedsl5 118111351tdeposited when1111811 thethe lakei11i5'1': 5218811stood .11at .1a higherE1i82181'le'1el.level. Tol'o tl181111rtl1th8the north the lake|11l18 beds LtILare masked11111511811 byt» more1111 )r8 recent:'8881‘1tttllL1\iL:nialluvium broughtlz'LttiLht 0011.11down by11'» theth8 river.ri1'81'. O181‘lL18l1ingLrOverlooking the 111111115plains and 511111111135swamps 011111851'131‘1'11i5the1JlLi‘101811111Sl1ttt11lJLtl8.1'i5111L_’11))10l'88tt1'J'Ltiittielesstltttrt5.000on the south is the old volcano Shombole, rising 3,000 feet to a little less than 5,000 feet.188:. In[11 513it8spite 1.1:of 811.1.5i11:1erosion that111.1: hasl..15' 81.11cut 11188115110115.precipitous, 1.1111.radially-arranged1l_\—.'11-1".111L1L-'-Ll 8:.111}'1.111~_;canyons on :tllall sides 01'of the mountain in the soft5' l". tLili'5'.tn.ituffs and 5h.1tt81"8d.shattered nephelinitesttL' 1:18iinit851.1l"11hi8hof which it i5is L't)111_1111.5'81‘1.t.hecomposed, the volcano-\'t)|e:1nt)- like profileprofile is clearly Ldiscernible"'e1'111l1lL .1tat .1a distance.0151111188. 'li181l.1tThe flat top11.111 111'of the mountain1118'.11'11'.ti11 is.'5 81111111115811composed ol'of flat-lying| .t-lyii'g tuffs:Lttl'fi5‘ tit'.1tthat 111'originally5: '11. iii-\' lay 111in theth8 wideHiLiL crater.811.181. Thel':1L' 8‘1"l'5'cliffs thatlh;ti 5'111'1'0L::1Llsurround muchmtieh of the 51111111111summit plateau'1'1i.-.t8'.1z Ooccur“Lii' 11Lnear' the111.: iinner'11'..'1' 81'crater1.181' '1wall,. .ll. (hethe flat-lyingll.tt—'._1i ‘.111'1'511l'tl'8tuffs of the 8113183crater 8\i1‘.’81'.t|}'evidently beinghem;r more11:1.‘1'8 resistant1'85i5't.‘11'1'1 t1.to ..11511.1:1erosion than‘.h:1 '11 the.he 11dipping11 311;; hL'Ll5beds 11i'tl'18of the 8111.81crater rim.:11'11. \'1851West ml"of the \H.‘Ngare1'8 .\_\Nyirokirk" '1plain"11'. is"5' :"11the 'southern"tthern 1:311tip 1'1l'1'heof the i\'.'li‘r\iLlKirikiti 1'1i11tl'111'1t1platform t‘L‘l'.(cf. Baker,Baker. 1958,I958. Fig.i'it'. I,l. 1'1.p. 5)1'1'1hi1.h1which is the l1.1'.'1'8:'lower .518;1step atof the L'L'111‘11'1115i1composite18 Nguruman1‘581111‘11111.1.'”1 8.581111111113111.escarpment. This lmxerlower 858.11'111118n1.111111hescarpment, which isi5 .1composed311181.188 L1.of basalts,.'5.1l15'. is1.5 l‘.high .11'11land 111'88i111'1L1L15precipitous 11.1to :hL'the 11111'1‘11.north, 1111'.but 511111825slopes :11at 1111151111111111811-\1"L-1'1'Llapproximately one and '11a l'..1l1half 1.88385degrees 511'Litlsouthwards'1.1.t.’'Ll5 11:11111'155'85and passes helbelow the 159118Ngare Nyiro.\_1_11'L11'1l11:nplain 1.11Lt'1hL'hOLildL-I'i'181i5'imiheand the boulder beds to the '1\1L.51west 1'-'.‘of itit_111.5ljust 51111.hsouth oiof the l’.tL_t:t.5iPagasi ri18:'.river. A\ i'1_‘\\few inile5'miles WL‘SIwest el'of '.l18the Kirikiti L:58.11'11111L'111.'.1111lescarpment, and just111.51 Litttsideoutside 1‘118the 11'85'11".1.1'8.1.present area, is the main111.1111 NgurumanNgttrumun '5'8'5'11"l-‘-"“'5‘Jescarpment1‘.: £11111and itsit5 southern5'11'itl'1L1'1181.11111111:'.1tit'\i'1continuation the 01Oi DoinyoDein_1'1'15.11101118~;1.‘.11'1_11111'1'.:.Sambu escarpment, both 01'of \\hi8hwhich form11.11111 the impressivei111111'L.5'.5:1L 1185:8111western margin111:11'8. 11 81of thehe RiftR:l'1 \Valleyl1iie} :'1:'1Lland rise1'i.58t.1to heightsh8igh15'i:18\885.5'1.1:'7.000in excess of 7,000 feet.iL‘Cl. 'l'heThe 5111111'nit5'summits 1.11'1of the KileuKilett r:ridge0148 :111Ll'15'1'1i1'uand Kaku '.1r8are .11at levels18'18l5' 1.1181111181111181:of approximately (1500111611006,500 to 6,600 feet,:'e‘t. and"11"8‘1 may11111:. represent1'8:'11'85'811t '.11'1an e\ten.5:011extension Niof the1.18 etLi—‘Lend-Cretaceous1"8t.188.1L15' 8:'L15'i1.t;'1erosion 5L.1"l.18.8surface recognizedreeogni/etl in the Namanga-Bissel\ztntttngtt— Bissel 111'8:1area 1‘.l1.1t.l.181't(Joubert 1957,W p. 411,4). T'1.8:'11.1‘1111'L"The mature valleys\'.1lle_\51~11on iliiall sides 81'of Kaku15.11511 hill 588111seem to1.'1h'e.1\have been grL'.LlL'Llgraded to111 a:1 levell8 . of.. approximately'1\.i1:1.1tL'-l_ 5.51.11215,500 feet,Feet. .1110and this011.5 811r1'85131111115corresponds with the levelslexels 1'1ascribed'8i'ilJL'Ll l1.by Joubert[81.111811 to111 5..1.—\l'1.1L1:1'.8sub-Miocene peneplanation11811811l1111.1:i1111 in the 5081south-westh— 118.51 11:111111'1118part of the n Namanga-Bissel\Lttndrr. -' 8l1'1:'8'1.area. The 51:11:111itsummit 1'L1l1g85ridges 111'1l18of the ‘11il1.5hills L811l1i_f.1‘.'18.l1.5Lenkijabe, Lesiri,.LL1l'L-0111'110..'111dtl18Loldorobo, and the south—11.851south-west sheLil'L'l8z'shoulder 11‘"of Losirual...15'i1".1.1 may111.11 represent1‘81 8.58111 1'8111:1.11-1t.5remnants 111'of .1:1an Lr115':1111erosion L18\'8lbevel :1:at altitudes betweenl181\.1'8-_:11 4.5004,500 and 4.7004,700 feet.l'88t, This 511111188surface '11.1.1'.1l1.iwould l.18be 1111111'01'i11151t8l}approximately 1,000i000 feet1'881 belowbeloxt' the 5Lil1~klit188nesub-Miocene SUE'i'LlCCsurface '.1a .5ht1rtshort L£i51111188distance 11.1to 1218the 8:151.east, 111101813;making it necessarytCCSSLtl’} to 111151111118postulate :1a Huhfault L 111:1111111'-l1:downthrowingowing11131513111i1111'1tel'1approximately 1,000l.1.|00 feet1'L-81 1111‘1'811851to the west 118t'1188:1between Losirua[11511811 111111and Kaku.15.81111. theThe gently 5i1.111ing.sloping Si:1t;:1Sinun plaingtlttfi: ‘1‘-851west 112'of lLosirua0'5'iz'ti.t L"\t811tl5'extends nitrth—ezt5ttt'itrd5north-eastwards along the 511111foot Oi'of the Basement 53518111System hill5'hills intoiii-:0 the Namanga-Bissel'51111':.1ng.t—B'558l .tt'L'.1area '.tnL‘.and 1111on into the t‘Cilarea west 81'of Turoka.Turning 581113141151South-east 111'of Njoro15111111 Sindio.‘51111li11 in the Namanga-BisselNL111'111iig.-1—Bi.5'58l 8118.1area the plain extends .15as 11a broadhair-.111 81‘1'1l_1:t_\'mentembayment CEISL‘Meastward'1.1 11111.1into H18the 111'.undulatingLit'Iltttittg Basement811581118111 S}'5itemSystem country '.trt'Jttntlaround Mamm—Matum- batu.balm. Thethe plain11l: 1: 1'. 811115i5'15consists L1:of sandy.1.'.._. .51soils.‘li 5 withh Ll.1.1'l1dark L'l'.1;.clay 51'1il5soils in311 51111111131:swampy '.11'e:'1.5.areas. Thel'he western

J -

part of the plain certainly overlies the basalts in the Rift Valley, and to the east they overlie Basement System rocks. The plain is due largely to the aggradation of alluvial sediment discharged by rivers descending from the high ground to the east. It is likely, however, that the sub-volcanic surface in the Sinun, Torosei and Njoro Sindio areas is an erosion bevel sloping to the north-west, the slope of the surface being approximately 110 feet per mile in the Njoro Sindio region. This erosion surface is earlier than the basalt vulcanicity of the Rifti Valley\' :1 3' floori ' and is considered to be of low to moderate relief on account of the rarity of01"}1‘:1:u1'sinliers of0" BasementBusx'm: System rocks projecting above the Rift Valley volcanics. The eastern or expm'u.exposed part' of‘ the erosion surface is at an elevation of 3,800 to 4,000 feet at Sinun, and slopesslope; upwardsUpr towards the north-east till it occurs at an altitude of 4,700 to 5,000 feet in the 'Njoroﬁorﬂ bhhSindio area. The vertical interval between this surface and the sub-Miocene erosion surfaceUrmucLis approximately 700 feet in the Losirua-Sinun region, and the lower erosion surface forming-.ur:1'_in the Sinun-Torosei plain is tentatively regarded as end-Tertiary in age.

III-SUMMARY OF GEOLOGY The oldest rocks in the area are gneisses, quartzites and schists of the Turoka Series, aL1 formation['01 which is considered to be of Precambrian age. They are folded on axes plunging gentlygcnﬂytuto the north-east, and have suffered regional metamorphism and slight granitization. TiThey contain such minerals as kyanite, garnet and graphite. There is no geological record of Palaeozoic or Mesozoic rocks in the area; the main period represented is the latter half of the Tertiary and the Quaternary, when vulcanicity, faulting, and later, lake and river sedimentation associated with the Rift Valley took place. The two central volcanoes Shombole and Lenderut represent the first vulcanicity to take place in the area, possibly during the Miocene period. These volcanoes consist of nephelinite and tuff, and trachyte and basalt respectively. Plateau eruptions of olivine basalt followed the initial period of rift faulting along the Nguruman escarpment, and are represented by the Kirikiti Basalts and basalts in the eastern half of the area. There was then a second movementmow, .1“ on the Nguruman fault and quartz trachytes were laid down as a thick series in Elmthe floorHum of the Rift Valley. During the eruption of these lavas a series of pebble beds and sands \3were laid down on the Kirikiti platform. The quartz trachyte vulcanicity ended with lhcfonnuthe formation of small ash and scoriaceous lava cones. Two series of lacustrine sediments were laid down on the lava surface-the Chert Series and the Oloronga Beds. The deposition of these was followed by widespread grid- faultingIHJL. whichI...“broke up the floor of the Rift Valley into a series of parallel raised and sunken blocks.Mocks. Thehe High Magadi Beds and contemporaneous beds in the Natron Basin, were dcpodeposited- in'1 two larger fault depressions and other deposits, largely unexposed, were laid down in the smaller fault-troughs. Following erosion of these beds alkaline spring activity in the Magadi and Natron troughs gave rise to the Evaporite Series which is in process of deposition at the present day. Kunkar limestones, gravels, wind-blown sands, alluvial and fluviatile silts and clays comprise the superficial deposits. The tabulation below summarises the succession in the present area, and the area to the north. The relative ages of the Olorgesailie, Shombole and Lenderut volcanics is not known, nornorjsitnn.is it known what the relative ages of the 01 Keju Nero Basalts, the Kirikiti Basalts and the[in-.3 basaltsL ' of the eastern half of the area are. There is strong evidence, however, that the basalt series are younger than the volcanics of the central volcanoes. The remainder of the succession is well established. The only firm dating by fossil evidence is of the Olorgesailie Lake Beds, which are upper-middle Pleistocene in age (Baker 1958, p. 37). Recent work in the area to the west of Magadi has shown that the Kirikiti Basalts rest on the sub-Miocene erosion surface, the inference being that they are Miocene or post-Miocene in age. In the report on the Magadi area (Baker 1958, p. 16) the basalts of the eastern side of the Rift Valley were tentatively correlated with the Kirikiti Basalts. It has, however, been suggested above that the eastern basalts rest on an erosion surface of end-Tertiary age, implying an upper Pliocene or Pleistocene age for the lavas. For the purpose of the present report the basalts of the eastern half of the area are tentatively regarded as Pleistocene in age, and as being younger than the Kirikiti Basalts.

\ 5

GEOLOUIE'ALGEOLOGICAL FORMATIONSHmnmtoxs IN]\ THE MAGADIMacaw): ARE-xAREA .xsAS .\A WHULI'WHOLE Recent Loess;LoeSR: fluviatileﬂux'iaiile pebblespelTtR Evaporitebutporite SeriesSELJCS (140[140 ft.)*111*"

UpperL'ppei‘ Pleistocene High MagadiMagi-iii Bedx'Beds and.and <:sediments-. snout; {atat Jiorthnorth end 05of LikeLake .. Natron f.(>40'1" ft.)

Middle Pleistocene OlorgemilieOlorgesailie LakeLuke BedsBail->1 (c.(1'. 175if ft.)1‘1.) L Lowerl ower PleistocenePleistoeene Pebble bedsbed: and sands.iztds ot’of PagasiPage (c. 100 ft.)= Oloronga02010111111 BedsBed: 1'1".(c. K]70 ft.)It: ChertChen Series tr.(c. Ft]70 ft.)I'm ScoriaceousSeorttteeous lava.;;\;i and ashestlies ofLi vents-.;1‘.'.% Orthophyre-trachyte0rthtitphyre-[rttehyte (c.(e. 150 ft.)it} Plateau Trachyte'I'rueityte Series (>1 I 1,000Limit ft.)1':.:- Lengitoto TrachyteTI"JC‘l‘1_\IC and 2111‘1trachyteYES of" Lengoralel. (> 75 ft.) 01Ol KejuKejt. Nero BasaltsB.t>:i‘:t< (c. 2452- ft.). BasaltsBusults ofol. the eastern halfhull oiof the S. Magadi\‘lilfJJ-Lll area I.(>I» 200 ft.)it.)

Miocene andand. Pliocene?Plioeene'.’ KirikitiKiriLiLi BasaltsBtimiits (>{511i}500 ft.)tit)" Lenderuttemierut Volcanics\‘oleanies (>r_ 2* 1,000111L113 ft.)l'[.j ShomholeShombole Volcanics\'e-!' L. i 5 I.(>133.:Ol'1’1.)3,200 ft.) Olorgesailie VolcanicVolcanic: Set‘ies’Series tr.(c. 3.7002,700 ft.)11.1

Precambrian Basement System (>t; 14,500[4.51111 ft.)[1.1

*The‘The figuresﬁgures in parenthesisparentlti~ E indicate' e the maximummaxim observed thickness of each formation. Where the total thicknessthickneSS of01 thethe formationtormattnn is1: known to:0 be'r-e ..greater than the observed thickness this is indicated by the §}'!'Dh(1lsymbol>. . tOnly101-i 500 feetfeet of KirikitiKlrllnlii BasaltsBasahs are3"e €‘.pOSe[lexposed ini:1 I'te’rSl]the south Magadi.. area. In the Magadi area to the north the Kirikiti Basalts are knownknmm totn C‘LL'CCLlexceed 1,50015111711130feet in1.1mm.thickness.

IV-DETAILSl\ ~DlC'I'AlLS OF()1: GEOLOGY 1. Basement System-SystemiTumkaTuroka Series The metamorphicmetamorphte rocksweb ofol‘ the south-easternSt'T-Llifl Item corneremite’ ofot‘ the .i“area forma a group"'01.: of hillshils which are the western and southern extensionsextenstons ol'of the MetuView and Luanjil t . 1ii hillshill) ofot‘ 1‘1ethe Namanga-\lum't‘ Bissel areaurea (Joubert,(.louhert. 1957).195?). The lithology ofoi' thesetitex: rocks1‘0;l isix very*- similar' to the Turoka Series asu.» described by Parkinson (1913)11913.! and asLL“. more11 precisely defined by Joubert (op. cit.). The writer follows JoubertJouhert in treatingtretttirg the TurokaTtirokzi Series. as part of the Basement System, though there is :1a suggestion\Uggehtlﬁn thatthut the TurokaTI Lu Series may1“ eventually be separable. from the remainder of the BasementBuaement System on the groundsgr‘otgmlx of differentt structureLire and;-1:.l degreeLl of granitization and may deservedesene promotion to a:. higher grade3.11t in stratigraphicalFiz‘utfgrapltezi] nomenclature.non-1." The stratigraphy of the metamorphic1:1ei;i:11orpftie rocks:‘neka‘ that[hut form thetl‘e Kileu-LosiruaKil-L‘LrLCSii‘J-J hills.. is not as firmly established as could behe desireddesu‘ed on accountamount ofnl indifferentimitti'erent exposuresexposure-s uni.and rapid.r < lateral variations\ztritetionea in some of the rock types. The following succession isi< establishedC'Sllli'Rl‘iCLl forlo:- theihe Kakuhiiktz and Losirualosimr. hills:-hill»: feet .5;8. QtiartzitesQuartzites of Losirual.(_13lI'-.l_l ...... 1,5001.501} 7.'.-'. Biotite gneisses.gneisses ...... 3,0003.000 f 6. Biotite-garnetBiotite—gamct gneime:gneisses...... 800SUD 5. QuartzilcQuartzite...... 1501150 4. Kyanite-biotitegneiaseagneisses . . . , , , 400:03 3. QuatrtzitesQuartzites 0fof Kileu ...... , , 3,0003_000 2. Biotite gneissesgneisses...... 2,2002 200 1. Quartzites()uurtxltes of KakuKuku . . 2,0001.0:}1)

I I'"'"'"]

Tht‘The stl.tt‘i:

00 - --S-o.. .

t==:~Muscovite -biotite gneisses ITIIIllJKyanite -biotite gneisses c=JQuartzites Scale inIn Mile‘Miles 0 lI 2 3 ~ ~ Biotite-garnet gneisses L—.__._._....L.__....._..._..l_—--n—-._—_lL J J j Fig. 1. - Vertica! section showing structure between Kaku and Loldorobo hills. 7

TheThc whole of the KileuK1:CL11‘11’.1gc1511L11211I-|ridge is built of thin[11111'1'1'C‘i.-'1‘-C1well-bedded quartzitesIIIE'IIL‘S separated'5'“ .. ILHCLi by[W narrow113'1‘1 1‘0-“51 21111113151impersistent5113111 b.0115beds 0'1of C‘I11.LZ——n11.15C01I.ICquartz-muscovite .51"schist. The formation marked| .' on the map should behe 11:15.11.regarded‘ as1511a 711111:zone in1 11.11I“1which 11,1111:quartzites"1?. 1C are predominant. At the southern end of the ridge 111::the quart/1113.5quartzites dip .11at 1111111115approximately11'11Ci_1-'101I50 degrees to the ‘1.C\i.-|‘-'1.‘I'west-north-west, but at the northern endC1111 they[111:5 are111C vertical1'CrIiC1'11 and111111 form[111111 narrow1111:'111I'I' wall-like' outcrops.‘ '."1‘-‘.15'. At-‘-1 the upper (western) margin 111‘ '1115 1111111'17it'C 2m11:1 111C "" of this quartzite group there are muscovite schists locally and these contain vein quartz .5C'gr'c'g11111I115segregations 'C‘UI‘IL1'lil‘iicontaining large blue kyanite crystals. A further quartzite group is found in the1111; western1111516 :1". .1111‘.limb11 0|of the Kileu syncline on the east slopes of Losirua hill-each quartzite forms a distinctive feature. Quartzites outcrop on Ropet hill and on the next ridge to the east, where they underlie J kyanite gneisses. The uppermost quarzite is a poorly-bedded very coarse-grained rock 11111111111comparable'1I.DiIL.‘ to111111111that formingI111".11111- I 111Cthe ridge1'.11_Ij'. of111 LesiriLI.‘.5 hill... These quartzites have grains one centi- 111Ctr1‘metre 111in 11111111C1-C‘rdiameter 211111and \I-.'C:1I1i".'crweather 111to 11a C11111'51'coarse 11:111.-reddish. grit.7 Two11111 C‘1'1111'5ccoarse quartzites,111: '.'.1'. .1-.I5. with-.I.'i.1'.'1 muscovite111115.111- flakes near the base, outcrop on the hills around the11‘;C north-west.101111-11135'. flank' of Losirua1.1‘5' and inI the axis5 of(‘1 theH... Kileu syncline three miles east of Losirua.1.115'1‘ At.311 111Cthe upper1:11pm contactC1111. they grade into massive21‘. LlRK ‘1 i' muscovite gneisses containing ribs UIIof coursecoarse 1111:11'1211C.quartzite. LoldoroboL11‘I 11111111111111hill,ll. Lesiri,L'CI'I Lenkijabe and isolated hills north of Losirua are also composed of1Il‘ 1.11111r17211'.quartzite, 1111-31:often 11:of \very1'1';.'1'-.I:1I.'coarse texture, resting in some cases on muscovite gneisses. Their position111151111111 in11"; theLhC‘ 511C'C'1355i1'11‘1succession isi5 uncertain,1.111111.' 1'1 but1112‘. they11‘. may._ be tentatively. 1 . assigned to a position 11111111111161111151'1'1'1i11:11111115515above the muscovite gneisses {1|of 111-.‘1I.111-I.'1C‘\.1...:I1i1the north-west flank 11.of Losirua.E .I~1:‘.I:1.

1:1(2) KYANITEK1‘1\1T1 (I151GNEISSES'1 Biotite311.1112‘11111.‘.gneisses with kyanite and muscovite:11'15C‘1I in variable. proportions form widespread and111111 easily€11.51i recognisable1'" _ ' ".C' marker111111'kC1' horizons.1 1'1/1'1115. TheyII 11:)- 1‘C‘C'11.'occur on1111 the' west slopes of the Kileu ridge, on1111 the north1111.“ and west‘.I.1'51 slopes:'11_11C5 11'of Losirua,L115i1'i.'.111. and on the ridgeI east of Ropet. In each case abundant' 1. uni-111'. kyanite11'j-.1:11.1C" float3111111 crystalsC occur' below the outcrops, which are also distinctive on the aerial' photographsphmogriiplh becauseIDCL‘. of the small-scale mamillated topography to which they give rise. On the northern1 . and western slopes of Losirua the kyanite gneisses are particularly well exposed.C‘I.pO.5'C-'.i. Where‘1‘. 11C. C:1c.‘-no 111..quartzite intervenes they grade upwards into the biotite-garnet gneisses. The71116 uppermost1111I1::'1111151 kyanite.I.\' 1111: gneisses. are massive, sometimes flaggy biotite gneisses, rarely with sparse1.1.56 small5.11111Il garnets,g111'111'15,'and.. withIth narrow1'I:'.“ beds containing pale greenish blue kyanite prisms51115 up1111 to111 half11.1if1111an inch long. MuscoviteK1115“ 11I'I 1.11 frequently accompanies the kyanite and clearly" replaces‘II’iﬁiI'LC‘5:1it along11111111: certainCc'r111i'1‘11'11Cr5.layers. TheT1: upper11311;" 200 feet of the kyanite gneisses are of this type-the lower 60-806.11 3111:feet,1.111.-.I.CII.-:1however, are massive poorly foliated kyanite-biotite gneisses in which muscovite is15 rareR116 1Iror 11:15C111.absent. Below 111Cthe massive beds are thin kyanite-muscovite-graphite schists and these111C513 overlieC11'C‘1'i'iC‘ a1'1 thin111111 boulderyL11.111]11'C'1:. quartzite.1;;1:1r. The kyanite gneisses are from 200 to 400 feet in thickness{111-311'1C55‘ on111 the111C northern1111111113111 and11111“. north-western1'11Ir1h-1. sides of Losirua and are cut off to the east by a fault1I:1.1I:-:[.-.:'1I11g.3l.(see Fig. 3). TwoT1111 kyanite111111121: gneiss beds11C11'5 1ICC‘I...“occur in1:1 ::1Cthe low‘11)” hills east of Ropet, and are. interbedded11::11I1C1:11C"1! betweenhcluCC.'-1 biotite§1i111i1C gneisses.211-3151111111Both containCODEE'HIH minor11111111." amounts111111111 of muscovite and garnet.1‘11-1‘1. TheT11C1111I1'1:more westerly1IC51C1']_\ 111of 1.111:the 1110two beds11C'd5 isi511":the more1111111‘ homogeneous11111111 11-11115 :I:.and contains the most kyanite, in theEh: form.111'm of0.'C.15i1\easily visible1'-15111.1: green:1 1'CC'11 prisms.rIri5n15 The kyanitek)'L‘.I‘11[C gneissesgnC155C5 that111111 outcrop11111:. 11-. extensively on the west slopes of the Kileu ridge are essentiallyCsscnuall ' similar5111111111 to111 thosethey: already:111.'.111_\I described.1i . Muscovite is a ubiquitous constituent and the ... kyanite’13:;111111.‘ content11111113111 decreasesdecrczwcs from the top11111 of0: the bed towards its base.

(3){31 BIOTITE.11711'1'11' GNEISSES1351151515 AND5.5' BIOTITE-GARNET GNEISSES TheseT1C5C rocks.C1C115 are.1.‘C best|1C51 seen5CC11 in111 the1111': core of011;the Kileu syncline and above the kyanite gneisses ifay- on"I11 the:hC northernnorthCm and1111-11 western\'1I '13.:1115-11‘:flanks of Losirua.' They are inhomogeneous rocks varying fromIrom light-coloured11I1‘11—C‘11‘1-1'.11 rC11 felspathicHim 11;: gneisses with sparse minute garnet granules to dark well foliated101131.811 schistose51:11151'05C typesBT11)?» withI garnets. in well' defined layers. The occurrences in the hills east 1I1of RopetRopCL are not11111 21111C11C1111.garnetiferous but1.111111121111111contain thin'1 quartzite layers. They are separated locally from the1111: kyanitekyunitc gt1cis5cz5gneisses below112111111151by mmone 1111'or two11111 thin1i1i:1 1quartzites. The biotite gneisses appear to:11 bebi: :1111IIIC11I"altogether absent31.152111 on111I the111C col1311'. one11:11: and.'11111 a:1 half.'.. miles" north-west of Losirua summit and at the11“.: north-western.’111. 111-111: r-1 CI‘LIlCend of1'.=I.'1C'the KileuK;!C' ridge.

. L

8I]

'll'LCThe lL_\L'.'Lkyanite'LC gLLaLsgneisses :LILLland Li'LCthe lLLL.LIiLCbiotite LgLELLLCLL(garnet) gliti\>c\gneisses LLLgCLlLCL'together L'CpL’CSCL‘L‘Lrepresent 'LLa LLCliLLCpelitic (}I'or HClTil-p'semi-pelitic. LC SIL‘LgCstage LL.withiniL.lLLLL L.La LL'CLlL.‘LLLLL:LLLLLlLL‘predominantly iiiarenaceous'L'LLLC‘L'JLL‘LLS SLLCCsuccession.SSlDl‘l. Tl'LCThe 'L'zirialionsvariations Ol'of lithology LLiLiLiLLwithin L‘ithis stage lL.L\C"have 'LL:very_ iLlLL‘likely l‘CLl‘.been LILLLCL'L'CLlgoverned by original \ilfi‘LlllOvariationsﬂS in LILCthe composiiioncomposition M LLL‘of 'LlLCthe .x'L;LiiL‘.LLCLLL:\sediments L'L'LLLtLfrom L...LLCiLwhich Ll'LL‘Lthey L.LL"are CC‘L'.'LCLlderived LLLLLland into :La lCSSCl‘lesser C\LCLLLextent lL}by gi'ainilizLLLiLLLL.granitization. f~ A group'LLLL'LL LLof lwbiotite gneissesC‘LvL L 'L'that ‘\is stratigraphically~ilL‘L'L LL:LLland liLlLolC-gilithologically distinct occurs ~1 iniLL LlLL.the CL'LLL:core LLLof LiLCthe LosiruaLULli LL L anticlineL..LLLLLL. .LL; l"L_‘l‘L“"-below theLi'LL': l:kyaniteLLl‘Lil‘C' gneisses.L_::LCi%~C\. TELL:The §_"'(.LLLCLgroup COLLLnconsists OlgnCihsLx‘of gneisses L.LLL‘.that locallylLL'C‘LLil) contain g.i.’"igarnetL'L LLLLLland LL'C‘are llll‘rl'lLLliilC‘migmatitic in L'LLLLLC‘L'L'LLappearance,LLLCC. irregularirt'CgLLl LL felspathicl'ClspLLLlLiC SLI‘CL‘LlLS.streaks, ~ \CiLveins .‘LLL-Lland .v'Llllzones L‘LL'of 'L:LL'L.'varying texture.‘LL’L.LLL C;differentiatingL C:'LL.LLi.L.:Lig LlL‘themLLL l'L'fromﬁ‘L‘. 'LZ’LCthe LLllLCL'other biotin:biotite gLLCiSSCSgneisses 'ClCSC'l'll‘LL‘Cldescribed LLLLL'L'LC.above. BetweenBCLLLLLCC them. .-._andL'l thelLL' LLLLLLL.quartzite'iLL LLi'LL'LLCabove LlLthemLCL.L is L1a ‘LlLinthin l'LLLLbut \L'lClCSDI'ELlLlwidespread bed of L'EL'L'LL—gL‘LiLLCLlfine-grained L‘LLssCLLLLC-lLLCLLiLCmuscovite-biotite KC‘.LL\Ischist :LLL,too L.LLLnarrow.L Loto 11C.be SlLLM‘F;shown on [Liethe nmpmap. 'LhercThere is .-La ~tj .‘il'I'LlI'l‘sharp LllHllllC.lL‘Llldistinction between the migmatitic' appearanceLL LCLLTL LLC‘C' LL:of Ll.L'3Cthese l'CC‘lx'Srocks Landand LlLCthe LLLLn—non-migmatiticLLLi:-_:LL 'L-LLiLiC C‘llLlI'llC‘lC‘I'character LL!of LltheLC higherlLLClLC beds and it is possible..LlL lC' LlLLL'.that ll'lL.‘\they representC‘piC‘SC.‘LL LlLCthe LLLLLLCL'upper partDELL". LLLof ;;a LliilCl'CdifferentLL SCI‘iCSseries LL:or SL>LLLLULsystem. ThislfLLL CILlLLLllL.‘l.question couldLl i.LCbe resolvedl'C\l.ll‘~C-'\.. lLLby iLCLdetailed:LilCLl nL'LL:mapping..Lp

L—lL(4) MUSCOVITE\iL'sL' L L\ L LL (.ixLLsuLGNEISSES a Muscovitel\Llli~$C‘LL\liL‘ gnggneissesTNC“ L.LLC'LLLoccur L.lLL.'LL.Cabove theLlLC upper'LLLLLLL' quartziteLLLL 'LLZlLC gLLLLLjLgroup L.onLLL LlL‘the LL'LLLLLIL—LLCS'Lnorth-west llLLLLkflank L.Llof Losirua.l.-.'LLiL'LLLL. They'l'l are.LL'C massiveIllll‘wrxl‘LLC‘ coarse-grained45. airLC l felspar-richj‘ 'L rocksL'LLLJL's containingC'LLnL; L'ing thin coarse LLLLLLrtzitCs.quartzites. SimilarSiELLilLLL' rocksL'LLClLs alsoLLlSLL. occurL.LCCLti' belowlLClLLL L.LCthe ElllLlf'l'Fllquartzite anon LesiriLC»: LL: ‘Lzl'Lland theLlLC LLtiLCL‘other LLoutlyingLing hills into 'Ll'lL.‘the can.east.

1 L5)(5) LIMESTONEl_l\lFSiiL\i A.-\ thinLiLLLL medium-grainedLLLCLliLLLtL-C‘LLiLLCLl limestoneliLLLLLLCLset in 'LLa ’eLClL}trachytoidl'._‘LiLl LLL.Lmatrix of intermediate or ...sodic'LC LLlLLCfinCplagioclase,LsC. minuteC' L'LLLlsrods LL'.of 17-.pyroxene'L_L\C. LC LLnLland LiLLLClLmuch ironirL'LLL LLL'C.ore. Inln H'f‘C‘C‘specimen 58/238 some ksmallLLLLli CLLlLCLli'LLeuhedra l‘lof .LLapatiteL.LLtC LLL‘Care Lisilvisible.l'LlL". Specimens 58/244 and 58/236 are fromLL Li'LCthe north-centralL. L"LLLrL-C-CLLLL':LL andLrid LLCsLwesternCi'n L.LLLL'Lpart L'CspCC‘LiL-CJ}respectively ofl LLCthe SL1.summitLLLL.LL ridge of the hills, .LL'LL‘.and :LL'Care ~LLLn'3'LLsimilarL' LL.Lto Il‘LCthe LL.andesitesLLL.C. 3185 Ln“describedilLCL'l LLlLOLCabove,. hutbut Containcontain :ILLllsmall LLiiLLLLLn-Lsamounts of a mineral of‘ lowZLLLL l'i'CL‘LLTLgCL'LCCbirefringence ;.nCand lowlou rCiLeLctiL'crefractive index in llLL-L‘the matrix.

~Numbers .L‘TC‘ll‘C‘Llprefixed by‘LL. 3‘458/ referL"li‘ toLL" specimel1;s-LLCL "1W“ inL“ L'LCthe regionallL.“__‘ L'LL‘LLl L'LLlcollectionsLI‘LL' lili'\LLof the; Mines\Linc? :L:LLland CLCLL'LLLCLLL'LL‘Geological Depart-l)i’ptl|"' rnent, .\..=rLL‘Nairobi, . . 9

T11i. PorphyriticPO1‘p11\1'111L‘ basanites11119111111681'L'1r111111.c111'L111'1i:11:11111.11'fxform the prominent dark hills11i'.1~' .at. the northern‘ 1 "11113111 margin1‘1111'1111‘. ofL11'1.L‘1111L“11Lenderut ~ and£11111 on011 the1111: summit5111111111 ridge.I‘1'L'1gL'. They1'11L‘_\ L‘L‘111111:1contain labradorite,1.1111‘.1Li1.‘1'11L‘. ‘.1.augite' .1:1L".and 1."11'1.11‘,L"olivine phenocrysts,1‘1‘1L’1‘11L‘rj1'5 a. 111Cthe latter11111L‘1'1:51111111'13111'3111'usually partly or(_11' completelyL‘L11111‘1L‘IL"1}' altered‘.‘.1EC“L‘L1 to112.‘ iddingsite.MG. ' ThereL‘1'L.‘ are.11'L'- numerous5'11:‘.1‘.‘L‘:'1.11.‘~ patches1‘L1'1L‘11C\‘ ofL‘1' clearC16111‘11r1111L‘iLL‘i11analcite in specimen511L‘L‘: 111:1] 58/237,51 :31. and1111.1 in111 specimen<11L‘._‘i111L"1 58/234511‘ there are‘ \111L111small ill-defined1.1-L1‘L1 :‘1‘1.‘. patches11.1‘.L‘1‘.L‘~ possibly1‘1‘Ssi1‘1y representing1"L‘_:‘.1L‘SL‘n1111g1111.111L‘LL1111L‘111L‘SL111.1«1.\.an alkaline mesostasis. t,J. 1.1111(b) 515.111.111.111.11"1‘Shombole 1Volcanics1‘1"‘...'1‘.1'L ‘ 5111111111011:Shombole \L)1L‘a.111L1volcano rises 11111.11'L‘\1111L11L"1\'approximately 3.01103,000 1cfeet 1.1"1.'.~‘1'L‘above '.11L‘the <.11"'L11;-1L1:11L1surrounding UL-VL:1111'}country 11:1L1and is scarcelyaL‘arCL‘h recognisable1‘eL‘01_'1'1i 11111L‘ as11< 11a 1111L‘1111L1volcano 1‘11on :1L‘LL1'.11‘.1account 1_11".11L‘of the L1L‘L‘11deep L‘1I1_\‘111‘.-1i1LL‘canyon-like 1111LL‘_\Svalleys C1‘L1L1'L‘L1eroded 111111into its sides.' 1C5. At closeC1OSC range,1111136 however,1:11‘11L‘1L‘1' 111L‘the volcanic1.11L.“.11.L‘ .x‘11'11L‘1‘L11'L"structure is‘1'.~ L‘51‘L‘L11'111}especially L‘1L"'.‘.1'clear-the-111L‘ 11:1:1L1L‘L'1banded cliffs. 111113111111:11101‘11L11111L111.‘illustrate the radial dip L11of I11L‘1'.1».1.\the lavas .1:1L1and 1‘11'L‘L31L1111L‘xpyroclastics “1‘and' 11131'1L'11‘12L‘111.1111111111L1L‘01'1L111'1;the horizontal attitude of tuffs 11ndand Ln11'111'1-L‘ralcsagglomerates that2.1111 fill1111 1111':the 0115111111original ‘L‘."L'tL‘r'crater isi~' 01.11.1115obvious. The .11L"average13C angleL' 111‘of L111‘dip 01-11113of the volcanics\1'01Cﬂn1cs 0111116on the 1.111111111111111111'111south and north riL1L1L“.ridges L‘:'of 811111311111:Shombole '1is 191‘1 degrees..‘L11'L‘L‘1L TheL- _prominent1111i1‘1L': hog-backL : .‘ " ridgeridgL‘ C115101'111L‘1111111111L111111111111east of the main mountain 1'.".is due 10to the1:1L‘ upfaulting111'11' of a portion11‘ 1'11 L"11 01'of 11‘"the eastern‘.11'111'21 flank1111111< C1of 1 the volcano.VL‘EL‘s‘. 811011111L‘1CShombole is1\ notable:1111111‘ 1c for101‘ the111L >1L‘L111'.L‘.x.\steepness L1;11\'of its '-.'1.'~slopes‘ andL1, the.'1'1L‘ ii'L‘LLL‘11L‘1'L'1'1treacherousﬁ 11.121111:nature of011116the rubbly1'111‘1111‘ lavas1:1\;151'11‘.L1and 11)1'1.1'L‘1.151icspyroclastics L‘1'1..1‘11L‘11of which 111131they .1:'1.‘.are L1_11‘.'111;1

1L(c)‘1 Kirikiti1‘11". ' Basalts The main features1'L";11‘.1rL‘.\ 111'111Lof the KirikitiKir'i1<.11 Basalts111151.115 11..'1L:have 1.1L.‘e‘.been described1.1L:"1"1.1‘."‘Li i1in Lt‘.‘1connexioni-1:".\.\ 'L111 with the Magadi\111g11d1111‘ca1B111LL‘rarea (Baker 1958,1‘)“. pp.1111. 16-18).113 1N1. '1'11-L‘}They :11'L‘are L‘\11.1:.LL‘.exposed along'1 ”51.1'11.‘the face 111'11'1L‘of the 111'precipitous lower1(1\\CT1'1(11'1'part 1.111110of the \gLJrLNguruman1.111L‘5L‘1‘11‘11111c12‘.escarpment and11111.1 on 1.119the lower11'1".L"1' Kirikiti1L::1'1'11L.1platform.1"1".1 1.111In the1311: present;)1'L‘~'.L‘.'11 area1111211 the ledge1L‘11gc (11'of 1.121"basalts111$ 11.111511which forms11.111“ 91311111:this platform1111 5.1.‘1‘1‘5slopes <1:‘111E111._1'L‘.

. I;'

10‘. 1'.‘

Little1.ittle 121111can 111'be 11:01:11added '11‘to Joubert'sL't'K 11LsL:'11.‘111..‘.'1description 111'of these reeks.rocks. TheyThe} .111:are charaemrizet‘.characterized by bouldery enterepsoutcrops and1:1:11' 1:'1L"L'occasionally11111 L"\11i'h'.texhibit '1'.a facies1".1L'1L" with s111'111small visible‘.'1sih‘11: felspar11315113111 phenocrysts.phenom sts. TheyT'he) 111'L‘11ss‘111‘1‘1'L-11are assumed to111 restrest 11directly1‘31' 111‘.on the B.'1.<.11‘.11.‘111Basement SystemS:...1e111 rocks.1'.0elLs Thel'he thicknessthickness'15is unknown but it is unlikelyunlikei» tot11e.\..'e1.'1‘.exceed 200 feet1'1.‘L"1 .11111‘1galong the L'..s'.L1'neastern margin111111‘ui11 01'of 11'1L'-the 011101011;outcrop; the lavas1111115 meare likely,like-13'. however,hottexLz'. to thicken1111L‘lLLn tov.ard1111' the nest.west. It11. is1s 01'of interesti111L'rL‘st to observe that the olivine basalts1111 1's eztncan be .1-shown‘.'1‘1‘11‘. to”11:1 be‘1‘: much1 .1'.L".‘. 3.1"11113_‘1:1'younger than111.111 the1111‘ KapitiK111‘iti ‘phnnniitephonolite 1.111111:of the Turoka511101111 at'etiarea (cf.(131‘. Gregory(1163301) 1921,1'; . 1‘11pp. 180-183).180— $31.

(e)1113,111"1‘1'11'l1Plateau Trachyte1,1111‘1‘111'1.15-1'1‘1‘1‘1Series The Plateau171111011. 'l'i'..L"113.t.‘K.Trachytes .11";are the1111: most1111‘1K'1 “111.1.‘K'widespread_.‘1'L11.1 hamslavas in the Magadi region and have beenheL-‘n fully11111} 11Ls11described'L-L‘Lt previously1‘12'111‘11 _ 11:111‘11‘1'(Baker 1958,1'151K pp.1111. 18-22).is: 121. TheyThe}; rest eonforrnahlyconformably on the olixineolivine basaltsh11s1ilts'in1n the L.1steast .1110and L11on the 1.1111111Kirikiti Basalts11.is.‘.lts in111 the west. The thickness 01'of the series isis 11111111011111unknown but11.11. isis L‘Li'tziinl3.certainly in51‘ ..1‘.'L'Lss1'.1t11'101I1t'eet'111111excess of 600 feet and isis likely111111} 101.11:to be 01111:of the order 01'of 1.0001,000-1,5001,500 feet being ti":l—;erthicker inin the west than‘ 1 in the L11K'1.east. The presepresent11 eastern margin 01'of the series is . 13103:”close to11') theth: former1111:111'1' limit‘111'11'1 of extension11s101‘. 01"11‘1éKeof these lavas,1111113. littlelittle erosion‘r1‘1si1tn hatinghaving taken 1.1111130place Sincesince their e115eruption.0110“ No'.\‘1‘ ‘1vents.-1“. 1' from111.1111 which1'1 they11111111113”may have1‘11‘.-.3 131111111311erupted 111are'1: known11101111 and211111 theythe\ are regarded as 1111cdue to 1"s11refissure e11‘neruptions' .."n~. for1'1‘1' this11‘11 reason.11.1.1111. In1:1 1.11.1131'13-11outcrop they;they are.re Characterizedcharacterized by broadly1710111111 5111‘spaced1e1‘. horizontal111.11 :1111‘1 partings1.11.11'1711 giving‘-‘i\111'_:3 thethe loose1111‘1.KL ‘1'1L'.‘.1‘11Lre11weathered rocksreeks ata 11flaggy1gg} appearance.appearance, '1'qThe trachytes113L111 1L.K are 1.11211dull greenish.- ‘ grey, 1'1'e111‘.frequentlyenth with1‘ it}: .'a sheen 1111:due to10 thethL presence of011111111many felspurfelspar miero—pltennetgmicro-phenocrysts.' The. a matrix. : isiK.‘ rough1'1; 11111 to10 tl‘1':the touch 12110and the 01113only megascopically111eg115e01.ie:111_\-' visible eenstitt‘entsconstituents 1'11are‘e 1‘1‘.1'erare _\1.'yellowishi101"1K11 1111.11.11alkali felspar151511.: 1.‘E‘L11Le1‘ystsphenocrysts up1:1‘ to 01one10 centimetre long.

In thin sectionSC 31101": the trachytes1.'.‘.L‘1'. 1'11'11'1'1‘1'L‘K'L111of the present 1‘1‘e'tarea 011.11can hebe broadly1111'11'11‘113'11ivi11e11intodivided into [110two varieties:varieties; the 111111.0101111‘1microporphyritic Loloitikush1 1 type131111: 1'representedL‘presenteL‘ by l'1'.‘1‘.11'four specimens.specimens, three from the Loloi-l.010i~ tiltushtikush plzite'ttplateau (58/230, 251, 252). and one from' :n the1.1LKsouth-west.'111111‘1—‘1-1 st side 01'of LesnitLesoit nearby (58(58/240),2401. and an "01'"orthophyre" variety,3 '..‘111'11which .sis 1.1016more 3.1.11.1es1‘1rei...1.widespread. 51111111Small 11111011aamounts 1.11of interstitial (11111112quartz 111110have 111..."been identifiedd 111 11L1‘1 Enin 1.11.1111both types.types, .'1K:as ‘1.1..1:was 101.:111found in .'K1.\111Lsome 111of the 11.10111ttrachytes of “1111-.the MagadiN'Izigatli 1:1e.11area (Baker‘ 111.11.“.1958, p.1'1. 19).191.

The The 1.010110“;Loloitikush. v~riety-‘._. is1K. 0 §)1__.11"1'.\1'11".:porphyritic ‘-.i.hwith 2 11.1to 5 111'liitnetrLs—10ngmillimetres-Iongs01111soda-orthoclase0111101311151: phem1er\"-"phenocrysts.' and green' aegirine-augite‘:“‘-:.‘11'1111L'- 1121L‘1'11.‘1‘1'1e.1'LtL1'\stsmicrophenocrysts 1:1in '11a tine-1.31111fine-grained:1e11 matrix consisting 111111111)‘mainly 1'11of 11'.alkali felspar prisms,' aegirine granules''..‘.11.'1L's .1'1‘111‘0'1and brown11111 1.111glasss interstitial 10to thetheLlelsp.felspars.11's. 90101101111111An alkali amphibole""1'.‘. pleochroic in shades of"011‘.L'-1‘1'11'1\.11olive-brown is ntniee.tble'noticeable 111in mustmost slides. Calcite veeursoccurs in V1.vesicles and in patches in the groundmass.1 111In one thin section 1158(58/230,230, 110111from the 1101111112northern 1301‘;end 111'of 1111:the Loloitikush plateau), there is much1‘1111‘1". t'inefine Lhaleedoniechalcedonic 111111117quartz in the 11101111111110».groundmass.

The111:11'11‘orthophyre1111‘113."'L \1.1IL.t3.variety contains'1‘..1. 1s 1:1Lthe Ksame'11:: 11'111'131'111K'minerals .1:and111 5011‘someL 1111of 11‘::e11‘11L'1‘11iFL1K'ptisthe alkali felspars are micfcnhcnomicrophenocrysts. The felspars'1'.'..ti1.'1'1'.K'011.11of all _generations .11'L'-are, 1‘11.1'1‘.e1'1'however,. Ktl1111'tpriK'ntsshort prisms with 11a 20111.zone 01of 1111s13"11e1dusty inclusions. near their '_'1'11L‘1L‘1'K'.borders. The matrix'..1r1..‘1. isi.K 1111.‘L‘hmuch e011coarser-grained{.7111tired thanthan in thethe L1'111'1itik11s11Loloitikush 13:11'1Ltype 111111and L'1.‘:“contains :1111'Lmore L111L11green aegirine-augite,: '.'.tigite. 0111111often in the form 01'of raggedrugged 111101'0111‘1L'n0eiwmicrophenocrysts.11K. Skeletal' '1 ophiticit ' kataphorite{LR-111111111111: '.K.is L'L‘111111L‘ncommon 111111and 01111117quartz 111in isolatedisohttetl elettt‘?‘clear gt'atinsgrains neetirsoccurs in :1111.'half .the' slides. Reddish brown.11 11:1to ~_:1'1.'L1‘.1K.“.greenish :‘11'1‘1'11111brown 13.111sglass t‘eetii'K'occurs '..sas intersertal wedges 1.111L‘11LY1-11..\1.1‘1'11-1C.‘1:L‘L.'\.throughout the slides.

Examples1'\"111_131. of the orthophyre variety of alkali trachyte were collected from the eastern edge 1'11of 11‘1L‘the Oloronga plateau, near the northern‘.L‘l'n 11111113111margin 1"11‘of the 1110.1area (51¢(58/227);227); 110111from two miles1111105 11..westst 1.'1'of Ndamo (58/229A); from the1.. west-K't K11"...side 01‘of the KiambuK'111111‘1: \'1111e_\'valley three miles east 0,1 east. of (")11‘11'11‘1101Oldomot (58/239); from one mile west.1‘.'.t 111of Ngadalat\1‘ 111111-21 1.53(53/249);2491: 110111from the LLedgege 01"of the 1FC'1C1’1}'1Ltrachyte 1111112'1‘11‘.‘outcrop threeL'L anti-L's11.1.'1‘.—‘.‘1.1:K'.1'L‘11miles north-west of ToroseiJDI'L‘SL1 ‘.‘.wellsL";1.K 15$58/261);1611: 111'1111111e\1e.‘tsi11efrom the west side of the SereatexSereata valley,‘.11‘L‘v three milesT.11L'K1‘.‘.".~1west .‘1of 11theL" K”south01' 1‘1 C1111end Oiof LL11\L‘1\'11"101'1”O151$Lake Kabongo (58/262);:{1_ 1; and 1111111from thethe plateau two miles west of Olonguruan wells.

The trachytes of the Loloitikush variety are' very11.13. .Kii‘nil'similar.11 to the 5011soda.111 11‘;trachytes1011\1135 01‘.of GiheleGibele typet\':'1..' described11LKL111‘L‘11 by. Campbell Smith fromlr.'11‘.'1 thei11L1L‘1.11‘1Kikuyu11 1011'.and KedongRedungz esearpmentsescarpments (Campbeh(Campbell 81011111951401.Smith 1931, pp. 220-222), and the orthophyre'.1'10111'1_.1'e ‘.1'.1Ll‘.1'testrachytes 1‘1ee11111p:11'.1‘-:‘leare comparable with the pctntelleritiepantelleritic 111101131135trachytes (01".(op. Cit.cit. pp.1111. 222-224).

It[1 is 011111:of interest. to111111111.”note that'..' the. rocks'1 ;'1"1r11‘.in§_:forming 01OJ DoinynDoinyo LengoraleLengorttle hit!hill in the \x'esternwestern part131111 1111‘of the 1"area1:11" rest directlydirL'L113. on. the..1.L‘ Kirikiti10111011 BasaltsResults and1.1.1111 1110are 11011111quartz 11'11e11ytes.trachytes. There is thus 11a strong parallel11:-111111121 with‘.‘.111‘. the 1.111.11111111‘.situation 111at the north111111.11 end 01of the Nguruman1\'gtiri.‘n'1.1n escarpment where a 11

Singic‘single 11'11L'21}".L‘.trachyte, 1111:the Lengitoto1 L113.11.1'.L1'11'1.L11_\1L1.Trachyte, restsL'L'stx 111'on 111-;the basalts.1.1L1~.1'.L.'.~. It1'. is15111"on this1115 L1'.'L1L1"1L'L':11L111.':1L'evidence that the 5cc011d11101€111e111111'111L‘second movement of the NgurumanNg111'1111111111.111111»faultis agar-1.11111.1511.1'11'.~__11.11L1.'"11.L'L‘regarded as having taken place .11.afterL'1 the111L111L'gi1111111g.beginning of01 the1111: plateau trachyte vUlcanicity1'L11L'LL111L'11) (Baku.(Baker, 1958,1951‘. pp.1111. 18 2111‘.and 1’1262-63).1:131.

1:111!(I) Ash 11110“.SL‘L‘1"-11'1L1.1‘L’.1‘~‘LLL‘LLL‘and Scoriaceous Lava 11‘1“of Alasho.1111 ()101 Doinyo[3011110911115110Alasho isi511101'.a low hill11111511141191.1‘situated 6.15".east 1.11‘of 8:11.1111111'11L‘.Shombole, L11111111composed1L‘.1 111'of ashes,:x'.1- 1:15.11tuffs and glassy SCOTiaCELHlS11'21C1'1_V1'C.111C1L111C1‘11L’131gscoriaceous trachyte, the latter being L‘L111111'1L'L.’.confined 1.1.11“to the upper‘ part..1111of the‘.1'.L' 1hill.".1 . The broad 111.1011open col bemeenbetween the111: L‘as1cr'11eastern ridge Lof 31101111111Shombole and Alasho is covered by yellowish brown ~1 crudely bedded pumice 1L111‘stuffs into .1a L1L'11depth1' 'in excess of 60 feet. These tuffs represent the first 1]}I‘OCILH11L‘Spyroclastics 1:1CL‘ILL.ejected 11'0111from 111Lthe \L111vent '1'1111'L'111which is on the hill itself. At the foot of the hill, and 1'01111111g11nforming an 011111L1'11outlier a 111111:little L115111'1L'Ldistance 1101111north-west. of it are reddish brown tuffs resting on the J‘14.. 11511135.ashes. T11LThe tuffs1111115 L'11111'1'L‘on the 111111;?hill grade1'.': 111111.116?upwards 1111.1:into agglomeratic beds and finally the upper part L'LJfof the1'1'L 1111113hill is LL111111L1SLLZ1L11nL'L1rL.1LL'L11.:Scomposed of scoriaceous brownL1: 1.1.1 11 51glassy' trachyte. This trachyte (specimen 58/231) (301181915consists 111of rcddighreddish brownb10111111113111_1partly L1L‘\1U'111devitrified1.111glass containing small alkali felspar phenocrysts. The Alasho5.1215110 101L21111Lsnz'Lvolcanics are L1."111L1'.11L1‘1correlated with1.1.11: 1 the11.1. ash and scoriaceous lava vents common in the ...... - -.'. 1.1. Magadi\111g11L11111LL1(bL11Le11958.11.area (Baker, 1958, p. 1‘1.25), which1:1'..are regarded as having brought the plateau.LLL_L trachyte‘L.L1L11_\L'L \LzlcamiLit}.vulcanicity 1L1to .111Lan end..'.L1 TheH1: yellowish11:1a11511111%011111brown 1151168111111ashes that L1LL1:'occur on131 111.:the Alasho-Shombole col appear to11.1 (51.1.1.1;grade northwards11L11‘1111.\L11‘L.1.‘1 111111into bedded1T\_L1L1C{111L1\1L1I11‘Lfluviatile 5131.1and finally1111.111; lacustrine..1.' beds, which are closely comparable"1:11.11'1111'3L‘ 111111with 1111:the OloronguOIoronga Beds.Buds. (1‘1(2) LACUSTRINE[.LL 1 8111M; ANDam" FLUVIATILE1'1_1.\1‘111'..1. 81-1'1213.SEDIMENTS The SCL11111L'111111')sedimentary dcposi‘.»deposits 111'of 11'1L'-the 1'11'L'5'L‘1'11present .'.1'L"area L'1L'1'L'1'closely 1".'L~L1resemble11'1‘L1 and are correlatedL1 '11with1111 those already mapped in111111LLLrL'L1the area 11:1to the1E1: 11111111north 18.11Lake Natron stood at a higher level,lL\L1.;111dand thcsLthese '.Li'Lare 11:11.1L'11121believed 11.1to 11L11be the‘CL'I‘L11L‘L1‘C.11“Lequivalents of11 111Cthe High1113:11 Magadi Beds. Additional information11‘11L1r111111i011 gained in111111L"the present1'111'LSL1'11LL1L‘LLarea makes1"11: "K'L.\1iit possible[111451131311.1111'1L“to give a brief account of the geo- logical1OgiL‘L‘Ll Lindand L‘111'112111L11i5101‘1'1climatic history 0111Mof the 1\‘1:.1Q.1L'.11..:'L..11

Nam; Nyii'o Ngare Nyiro Basin13.15111. Magadi.'\1."L1__'.11.11 111311511trough I Legemunge-Koora1_L1_.'.':1'1'L. ..1 trough1"

High Magadi1\.1‘Iug1.'1L'1i Beds 01‘of the111L' Evaporite1-1. Series. 11'. NatronNatlmn basin11115111 1111Land High Boulder1;‘.1:11.."11.'1L'1 111::1'3.trains :"'.;and h11i_1\1\'sands terrace High- ' Magadi' Beds ------Ngare\gm’c \11-110Nyiro BedsB13118 1L111LL=(Lake beds,1L11~. 11:11:1'L11unexposed but Olorgesailie(31.0 Lake Beds 111.011111'1.probably_1~ 1occurring in the Lesoit,it?“ 11"; Sereata, Narsur- :11111'11131L315ana valleys LLLJ:etc.) "'J ------, Pebble beds 111.11and 51111L‘1nsands (Raga-1S1)(Pagasi) r (MorongaOloronga BcdsBeds and the111C .1511.ash Oloronga('J11'11'L'111L11; Beds1'11t beds 111'of AlashoAinsho ChertC'1'. SL11LSeries --- 1______----- Rift Valley\"L111CV' 1:11:15lavas (111051111L(mostly quartz'1-.".1111 / trachytes11;:L'11_11Ls.1.111.1and 0111.111:olivine basalts)1.1.151111~ ------

. ------.. 12

1'_('¢'_1(a) CI1'H'1'I'Chert IS-E‘E‘I'L'A'Series The principal[‘1' inLiITII' outcropsLIt' iL'I'l‘1‘-I\ ofIII‘ the".IIL Chert('I'IL‘I'I Series are in the vicinity\1C111'.1}' of01' Lake Magadi,MagIIL'li. onlyL.‘111}"I'\311i.111small patchesI3131C1‘CH beinghe 115: seen in121'1Cthe presentII:L ent areaItI‘eII-I nearme itsIts northern margin between OlorongaandOlot'onga and Rekereien.1~‘.I:1LL‘IL';LII. Thei'fte cherteI‘IL‘It bedsL ILIII invariably.II\'.I ‘.ijI rest LiiI'eL‘t'I'Idirectly onIIII lava1£1\L1 and are variable\Ltz'iLtble both in thicknessthiekrtesx and'III‘IL‘l lithology.|itho10g_\'. ChertsCi‘IeE'ts ereare C\]_1"I}?‘IUL1exposed onIII-I the crestseI'LIsb ofL.I|' the low10“ escarpmentsL‘SL'I'L'II'prI‘IerItS along the motorIIILILUI track north-west11L11'11‘1—'\’-L‘S[ III'of Rekereien. Often the 0r]1_\'only irII-IiieILIItiImindication 01'of these beds is the presence 01'of L1a gravelLIrtttLe1 I'II'LI1IIL'IIILof chert L‘LIII-I-Iti'ting.rcontaining castsI: 01'of Viviparusl .shells11; buthL:I e15etxfter'e.elsewhere, sttehsuch as on the brow 05'of the ﬁfirstrst LSCI'IrIIr.escarpmentII west. - 3 of".he13III-1 (,)1I.I.:IoneII r7.. Clay,. '.I with honeycomb structure . . OlorongaE :1 e I 1I.IIIIII'- 1 'O'It DI'UI.ZtI \i' :I. L,'1'Ll\ .. 28nix Beds “" i.5.6. Poorly bedded1 ..L1.:IeL'I1L-Liyellow-brown5111\‘siltyI. IIhclay.L1IIeI'IIIIIII-J'1II\II. . . 1,L 5.' Yei~~~~~row~ ~ell b~~ded ~i~tswit~.chert.~nd la~~ 4 4.4 Nodular kunkarIiLI'II‘ limestone11111L‘S-[0‘1C ...... t1 _: 71 1'" 4 (..ChertI. p. Viviparus chert "a. I' .~ I.2. Pale1’II .' green-grey conglomerates I'.c.5 ISeries- 1‘s . - 'k L 1Lt.1. Like"Cherty silts and clays . . c.1O (QuartzIif}t.II-.:t trachyte lavas) In1II. ti:the face'=-L'LI IIIof the lowIII-II. ea:escarpment1' ‘:.‘T.L‘!It I1'..I.tthat 1'011115forms 'IiILIthe eII>Ierneastern marginInnrgtn 01'of the OlerongnOloronga I11IIIteII'IL.plateau atIII the pointIIIIirIt whereI-.I. here the motor-trackmate-r II‘IIL‘k CrL‘HVLN.crosses, is :I.a further|'I:I'ther seeIiIInsection in Cher'tChert Series merhIinoverlain byh) sixSE\ and a:I half11:111'1'eetfeet of01'01I1rtmeuOloronga Beds:-BLIis'. feet'LI'I' ILLIJII'IIIg'II(Oloronga Beds)BLL1>I If).6. BandedBun-Lied. buffhtIII' LII'or paleIIItle green.. en LEIL-II‘Ichert \Iwithith Viviparus1'I'"|'I'";'7I'I'1’I'I’.\' castscasts. . 3 5..5. Pale 011\Colive green volcanic\I.I.1e;'IIIiL' L‘hI}clay ('.c.55 4. Grey \II1LIIIIII'2LIvolcanic nIIIe1gravel 1!-21k: .1.3. (I1C‘C1I-"Green-grey"- \1.:silt ‘1:4 2. T111111‘1;11\'1~IIL121(;1|'.Thin platy kunkar .. i 1. (kg-greenGrey-green friable1"Ii;I'L‘.-1e \I.I.1e;t'.IiLvolcanic xiiIsilt 5 (ISL'Lj'1.'IIi_'Cl‘Il.N(Scoriaceous trachyteI:'..e':1_\:e lava)..'-I L11 InIII. the immediatei eIIIIIIe \vicinityieinitI -of.1' HmtheI IIIwesternextern \L3[set 01'of hot springs at the northern margin of1t1of theIe area.area, and formingion-tine :1a series LIIof lowiII-I'I flat-topped11"1-LUDPLIL1 hillocks1—1111{FC1\'N toIII theEhe south,south. ELFCDQL111\IC\DOSCL1are poorly exposed Chertscherts e'LIIIII-edcapped by1‘) kunkarL'IIIkaz‘ limestone.11173‘I‘I'121L‘. The1"1:LI plain"§1[I"L1.11.LJ'11"0surrounding thesette hillocks1ii|'.I)LI'ILII consistscone-.515 ofL31 High11iIIh MagadiN1 Igati. BedsReds L1eIIIJsitL-L1LIn;deposited on an uneven surface' k. of cherty(I IILIL1ISIrocks L'i1L1and tnethe 5LIriIILIesurface 011118131'0UT1L115of the ground is littered \Iithwith :I.a nIIgravelL1 ‘31of eitL-IIchert IIIIplatesI‘.I es I.derivedeI. I-IL-I'. b;LLU\1Ui1by erosion. These ehertgcherts nIeare probablyprebabiy simihtrsimilar to those described at the south LInLIend ofIII the ridge e;.steast of the south-western lagoon LI1'1_.II1I'eof Lake Magadi. (Baker 1958,1958. p.I). 291.29). 13

One mile due north of the junction of the Gelai-' andu'u‘. OldomutUhim“ IL motor-tracks,|'1L‘[l‘-|-'\‘Llui\\}\ north-west'1<\"‘.|‘ ssh-I.

of the Lenderut hills, lake-beds with cherts are exposed" in aL1 gorgeCL rc leadingkudl. - through' 'amg'n aat longIn . narrowijLII‘E'L'JK'. fault-block. of lava. The‘ beds'lt-JUL dip at three degrees'c-cs. to10 the1h: north-westnm i'i-‘W'L’hL andLil'ﬂ. areLil'C confinedccnﬁncal to[1.) ;la narrowI1;:I'[‘1'1-'-. fault-trough; thelllL‘ HALE;succession is as'.|:\' follows:-i-._\'.|;1\\\.' feet (Unconsolidated chert gravel)

-1 Chert gravel cemented by kunkar" limestoneIiI.‘1-;~.irtnu ...... Il L\J Chert bed-in irregular rounded lumps ...... 3 , Tough pale brown clays ...... 8 j Yellow-brown fine silt...... 18 Lava boulders in grey earth...... 6

. l . ‘ . .. .‘ | .|.. x Lithologically the outcrops of Chert Series in the present area'n'cu ClUrq‘»closely resembleJC:C|;1bLC those:I.[).\\. already described for the Magadi area to the north; the rocks are clays and silts of volcanic derivation and the sequence frequently culminates with a fossiliferous limestone which is usually silicified. The patchy distribution of the series is due primarily to post-depositional erosion,:. u l. ail large| part of which probably took place before the deposition of theI'm: Oloronga()lo‘mngd Beds,Beds. since)i'JL'C the1.“ latter are frequently observed resting directly on lava. (b) The OlorongaBeds The Oloronga Beds were originally established on the basis of poor exposures on1 the1.110 Oloronga plateau in the area to the north (Baker 1958, pp. 27), but much more satisfactory.. m'_\' outcrops were found at the southern end of Oloronga during the present survey. At.‘x? theI“: south-eastern edge of the outcrop of these beds on the Oloronga plateau, just northh of01' the motor-track, six and a half feet of bedded brown silts rest on theihu Chert(fhct't Series.Scricg. FurtherFunhm- north on the same escarpment the Oloronga Beds rest directly on lava, havingm‘ u au thin“ ' 1 lavalaw conglomerate at the base. At the northern boundary of the area the Oloronga plateau has a smooth surface of kunkar limestone sloping together with the beds it overliesL‘- at' l twoint) degreesLIL‘._L'.I'L‘\_':~ to the west. The Oloronga Beds together with the overlying kunkar pass' beneathhut-neuth the later121m- sediment filling the Ngare Nyiro basin. Between the Shombole motor-track and the northern boundary of the area the Oloronga Beds are exposed in a series of low hillocks, the unusual dissection of the plateau being due to a river which formerly!'|_\' LH'JZHLE-‘idrained :hcthe low-lying1L)\\-‘._\".'_1<' ground east of Oloronga and which cut through the Oloronga plateau:m and11ml endedL‘nc‘ud in theLin: NgareNL‘ Nyiro basin. The lake-beds exposed total some 60 to 70 feet in thickness and are homogeneous light to dark brown silts showing little or no bedding. Occasional gravel layers are seen in the middle part of the sequence and near the top is a resistant porous clay. East of Oloronga the westward-tilted lava-blocks are covered by a thin layer of Oloronga Beds and kunkar limestone, but exposures are poor and it is only on the two low escarpments overlooking the northern end of Rekereien that useful sectionsu ';~; are found. On the face of the higher of the two escarpments between one and three-c milesm. cs northnm'L'n {)I‘of theIll-.3 ShomboleShombolu motor-motor— track a number of sections show Oloronga Beds resting on Chert' " _ SeriesSo: in rocks.rem-1w. TheTltc two:uo best5-34: sections are as follows:- feet Oloronga Grey silts and volcanic sands. . 10 Beds {Brown silts with rare lava pebbles 5 Massive olive-green chert. . 2 Chert fLaminated olive clays and siltstones 12 Series I Pale green to buff cherty limestone H I l Yellow-brown ripple-marked silts 8 (Trachyte lava) c.10 I Oloronga Brown silts . : .' . . . . . Beds Lava conglomerate III SIlt . . . . c.2 {Yellow-brown coarse sands withiz‘n clay lenseslam-:5 .. _ .. .. c.8c. 8 Chert Series Viviparus limestone 0-3 (Trachyte lava)

. - 14

A.-\LiLt'LL:1L117-Llittle further to the“'north-east,'L-11‘.-L‘LL L. betweenL'Lr‘LLLL'L'L' 11""the two escarpments, is the section of Oloronga Beds71cm and C71'L‘L”7L‘.~7.CL'7LL'Chert Series tabulated27.1 1717.36 L111on 1‘1.p. 72.12. Here77:1' 31 feet of poorly bedded brown silts with a thin171111 CL.1:1g71..1111C1'conglomerate.L' 111at the1771': 7.11113base rest7'L.7\7' L‘.‘on'L L.an‘11 L.eroded1' 1L'LLL'7. undulating.111 L71.‘ 11.11.11: 511131102surface 07of Viviparus7171111117?“ Ch'ert'.chert, thCthe latter7211121 being.1 L'r CL'11L'1'L'L7covered byL1_ L1a layer7.3. " of chert.'L pebbles11.17175 LL111L'11'L‘L'L7cemented 7.11by kLLnkLLLkunkar 7i111c511'111c.limestone. The 71Cdsbeds (.7117dip €177|\Cat five dL~_CLL'.‘L.'1degrees 7toL7 7.71“the 717wwest.17.

The ()71”11'L111___"LLOloronga Beds1:71:17 01”.”:are also exposed west of Loloitikush'1-1‘1121Lux7'1 and north101711 <17of ()701 DoinyoDoinvo Alasho,AlLLs'nn whereLL'iCi‘C theyth-::_1 dip gently‘ north-westwards towards theLhC 'Ngztt‘cNgare NyiroNy'im swamp. Four miles11117CS north1101771 of07 01(_77 DoinyoD‘L_"11".'. Alasho five feet of brown conglomerate'L1111C1'411C C(.111.”~:istingconsisting of rolled 11371ash .11.;1‘1:fragments11C11ts an.are cur-Maiexposed in streams,. and at intervals down the slope to the swamp. One mile further721117.711: 1' northnot theL71”: Shombole“17111112171L'L7C motor-track1”.1‘”. ‘ crosses a low ridge 117'of _"€770\‘.7’S71yellowish green clays and silts51715 dippingL'1'7.‘ at three degrees westward."' All the.'LL' C{:‘L7‘:¥'77'L‘\exposures of111' Lh1C5Cthese beds are capped b\-'by :1a ‘1 L.7. thin[11111 kunkarkLL:1 limestone7777:71777..7.7‘_7 which'77‘L'L‘7 slopes'\7L.77ﬂ..7‘-L north-westwards.771.7. L‘LLILLLE.

ThereThem LLt'L'are no11.1 exposuresC\.1‘D:‘.LLL‘~” overL'LLL‘L' the greater partL of the “7111113slope 07'of the171C broad {'01col twccnbetween 0701 DoinyoDoin} L1 Alasho.-\.L'.L\7'LL”' andL121..i the1h; L'LLsLeasternC711 part11-”...of Shombole._. T‘..rLL'Three miles17'11'7138 north-westnt11'171-west 01107.of 01 DoinvnDoinyo AlashoMasha 51:11.11"shallow stream-sections.2111—1 _‘77.‘17‘\ exposeC‘. four to six' feet' of'. 7'Cconglomerate1111111 11e1L1LL consistiaconsisting t1?of L'Lshash fragments77".7L77T1U‘1TN‘ andLLnL'. lava pebbles up to a quarter of an inch in diameter' ' .11in L1a 7.1;brown'own 5.17711silty matrix. SmallSm; 77 oyster77‘\\' shells occur in thes~ beds which, towards. Shombole, (_11”gradeads 71110into CCL11€11TCL7cemented pebble-beds11Ch7'17: derived from the Shombole lavas, overlain.‘ 31711111 by71‘ L‘L.coarser111131: 1’ 700513loose L1Cbb7CSpebbles 01‘of 111mmmore recentL'Cccnt LL.-date. One. mile north-west of 01 Doinyo Alasho\7 L710 5.streamL'LLC 1. :11 gullies14177765 Cxposcexpose 11:77well bedded brightbright_\C77L1‘..'—"L'L.L'1'CL‘LiCthCyellow-brown ashes which can be seen to underlie the red-brownL'CL7. 1111111111 I'd-7Htuffs that171' 7.1. form711.1711 thCthe lower71.1wcr slopes5'7L'1g1C-s L17of the171.hill. These ash bedsL'-' cover.:"‘LL'.' the111C 171073whole of071716the sL'LLiL'77Csaddle betweenbLt\\L'C11 A7ushoAlasho and 871.111.117.107;Shombole, L711.1'_.11'11dipping gently north or north-westL'.L.11'L'.1-.' 1:11 1.11to L7'1Cthe 11111171north 111‘of the sudmcsaddle and dipping 5011171south [1'1to the171:. 5south.111 ‘71 L.1.7of'7:it. A good river section”.'L' 71111. -.1:1on ”.71:the 50.1171south side threeL711'CC 11177L'x”miles $(3U77‘L—WCSTsouth-west 0.77:\7E:157‘l()of Alasho yields‘1C7d8171‘Lthe following1L”1711.“-.. section:- Jr‘s...feet L Kunkar-lava gravel 1i Hard buff gritty earth 711i l1 Platy kunkar limestone ~ Brown earth with rounded pebbles7.7::- of111' kunkar}\1:1'.7< .1' and‘.'L.1‘..L7 lava7:11L1 .. .. (16 -erodedL'1'L.1L'..‘L contact-.1L :———. Homogeneous7-711”'L11L'L;:'_‘ brown unbeddedL. 111CL7..lL'L‘. pumice'.'L‘..""-L'L; tuff‘. 17"2' ...... 7545-5050 The ash and tuff beds were derived from”L1 77701 DoinyoDL”"L1‘}”L1 .“L7LLs71L1Alasho LLnLland 111111311”appear 11.1to 71.116have been deposited sub-aerially in the vicinity of that volcano,”' ”. LL.1.”Eand 11.1to pass11.1“ 7:11.11into thCthe 7:1L'Lhtr711Clacustrine and fluviatile reworked ashes and finally into typical Oloronga..LL BedsBL'Lis 71.17‘77‘101‘further 1:1)to 171::the 1101171.north. For this reason they 411-.11‘.are equated with the Oloronga Beds.

Oloronga Beds outcrop immediately south of' 01UL DoinyoD-L1.:1\L1 LCngt'nLLengoraleﬂC :11at 1711:the 50111710111southern end of the Kirikiti escarpment, where they occur banked' against1111 theLhC t1'trachyte'L L'hyie 01of LcngorztleLengorale, and are overlain by the pebble beds and sands of the Pagasi. They7hL_\ COconsistSist of 171Cthe typiCL-Lltypical brown silts with rare thin gravel beds. Bedding is poorly_ shown,. L t . but 711C;177_\”locally 1711:)”they SL’CITLseem L0to (.7711dip steeply to the east and are affected by minor north-south1171 faults.Emits It71'1'1is 71CC'L'LL15L'because 011711:of the Faultingfaulting that it is impossible to determine the thickness of the 1..succession.L':L'L.'wL'1_111 L'LL11L15CL7.exposed, 7.1111but itit 1111151must CXCCCdexceed fifteen feet. The occurrence of the Oloronga Beds resting 11:1on the1;1C Kirikiti747117-411: Basalts7311'1715 is7.5 117‘71111101'12111ccof importance in so far as the dating of the movements of the Nguruman.LLL11'L1n fault1111131 is'5 concerned.con'L'L'rnL'Li. 7711\‘111111761'13This matter is I discussed in the section dealing with geological.L'-_z'L'L17 history71311111 (p.11.1. 20).2171.

L‘L‘LL' P LL'LKL’XL II, (c)“ Pebble17"L7L‘7 ‘.L Beds17JLL‘L.1 L.7‘7LL7L\.111"L‘LI\and Sands (7.7of the Pagasi Unconsolidated pebble beds and sands occur.__L'_111' L”11.1:'11L_'on the Kirikiti.\'7L ' 71707111101111platform \1151west and north- !~ west1.111257 of01' 011.77 Doinyo7'} Lengorale and are well exposed:L7. in1L1 t7LCthe gorge3.:L11L3C cutL‘Lit by711-171Cthe Pagns'Pagasi 7'7\CI'.river. They restE357C‘7111‘11_1.11‘1.;..on the upper surface of the Kirikiti Basalts and.1117 reach'.'L=LLL"71. aL1 maximum111L'L\1111-.1n1t71it1Cs_.thickness 07'of approxi—approxi- mately11111L"-1‘ 10077.71.) feet,L'L‘L‘L. thinning171.7111" towards the north. BanksJL 117LL L'L7LL1L'7L1L17135of pebbles rest1C.”t 1.111on theL711: O71..11'1111gaOloronga Beds and surround'L'L'1LLnL:7 L'L‘L1‘L'L'"L.1L1LLoutcrops L1.of the Lengorale. . trachyte.. "7'The1L1 1.1L'L.11.17L15pebbles .1eare'11C77well 1'0L111dCdrounded and consist of111' Basement13:151.:11311t System313151 quartzitesc1321; and pegmatites,'_ _. ‘ . - in1:1 a1'. matrix11117311 ofL1.\111L.-.sand. BeddingBCdding is CKtrCmeLextremely 15

,y..1 i‘L poorly developed:m; andm.:.11'many”M of‘ . the‘. _‘sandy beds are sandy earths. These beds togetherL with‘ the‘ basaltsbbalis anon whimwhich they1:19;: rest are cut off by the Nguruman fault, but much of theam monggxluiimcdunconsolidated material177311-JI'iL‘,i hashm beenbum sweptcm over) u‘ theLilo escarpment‘ by:_. the' L PagasiX ‘ _ -i runriver and :uucpomcdredeposited min ilHLlVlLilalluvial fans. The pebble beds and sands are of fluviatile deposition and.4 waywere laiddid Gov-h.down (mon incthe Kirikiti platform as a result of the rapid erosion of the precipitous. ms: Nguruman\ 1. cagmpnm'nescarpment "; [v.0two nu,miles to the west. Similar accumulations‘ of ,pebblesotm ' gm:and NEWSsands \were laid down\21 {:1in thcthe Vicinityvicinity Mof the Oloibortoto gorge further north\ri'w UH;during”mg in:the cmptm‘,eruption Hi'of L316the KirikitiKimkiti BasaltsHmultx , (Baker.(Baker, 1‘3,1958,.7 p. 16). (d) Sediments in Fault-troughs The sediments described in preceding,; wqciaanssections mgr;were Ladlaid Cmmdown previous1w, mug tom themg «.pmxmepisode of grid-faulting which broke up the Rift Valley'~. gm: floor.i‘n’mr. A:\ furtheri'm'fhcz' gainseries \‘1'of scdinmnix‘sediments \m'cwere deposited in the depressions formed by this faulting. Most of the larger fault-troughs in the area are floored by sediments, but itAL isA com-[0111— paratively rare that they are exposed._ Study\ of‘ stereoscopic‘2 pairs of air photographs_ L.}‘}‘.i nhmmshows clearly, however, that those fault-troughsm ‘ ; that’1 haveq no.w outlet0111a for1171' surface*»lli']m§L drainage arcare partlypun!) filled by lacustrine or fluviatile, sediments,U and, it isx noticeable that. 1.1in mm:some gmjx'cases the1h: troughs[11‘- have been filled to the level of the lowest part of the surrounding escarpments,W1: 1 and have then overflowed into adjacent troughs at lower level. This is particularly Q‘conspicuous in

the case of those troughs crossed- by the‘ r . ToroseiAr ,‘ and Turoka rivers. Light-coloured clays and1 siltsw w with occasionalw thin gravels are-: cx’gvoxcdexposed fa!at the111: wuLsouth-westfi—wcw side of the Rekereien valley and in‘ the1, centre, ’ g ofi the Lesoit trough. Elsewhere'. i611 0115:»only red chi)clay soils, light-coloured dusty soils and, in swampy“ areas, dark clay soils occur in'1 the flfloorsours 01'of the troughs, the largest of these, being7 the Oldomut basin, which. ”‘1 maoverflowedfia» 311m:in the fpast into‘ them; south-easternRummy A L armL of LakeL‘ , Magadi.‘~ The Kibangaini depressionMn‘aslul': “miand fltheic Em;long Sereata- hmmu:Kiambu xvalley:1; are similarly floored with «sediments. Further1 m mto Lik-the maxi.east, non1111;the mixmolivine basalt,Ktxuli Gun!"outcrop, the:11 soils in the troughs, n are generally, » brown and :‘Ii'jlwmglutinous when‘lun wet,gm and {L1Hfull 0i-of basalt@1q boulders,:H; and it seems likelyA , that fewL, if any sedimentsm wereM; ‘urcvmjxxitcjdeposited 3non thethe basalts.MM“ fThe Nasurana,‘ Kiselib,k . ToroseiV ‘ and‘ Olorguruan,1 basins and valleys are in a different category,gnu; W}, .for they haveK allL“: been_, traversed. ¥ by, rivers2 originating on the high gneissin ununlrj‘country [uto 1h:the QMeast 'of the RiftA Valley.W TheyJ x arel partly, filled‘ by poorly bedded fluviatile quartz-':I'I.’ \nm’nsands which M»:are A!‘of comparatively.‘uV'm; recenty ‘7date. SimilarSimifl' sedimentary7.“ " infillings‘ '77 of‘ V fault-troughs'7 , occur in the Magadi area to the,L‘ north, anduni in: , "the absence of evidence, to the, contraryv * ‘ it was suggested that they are the equivalents: of 1h:the Oloronga Beds (Baker. , 1958,7 p.’ :727). Although the Oloronga Beds may be preserved:‘ocrtul in [11:the deeper fault-troughs‘ , of the western part of the area it is now considered.cd thatthat [hathe unexposed lake-beds confined to fault-troughs} , 71 are1; almost certainly later than the grid-faulting*J'jL‘C-ilauit‘ ". and must be correlatedA with the, Olorgesailie‘ :9 "Q Lake‘ Beds and the High Magadi Huh.Beds. TlicThe uppermost beds in the fault-troughs are almost certainly equivalent to the High Magadi Beds, for they were clearly deposited after all faulting and tilting had ceased. Some features of these sediments are discussed below.

‘(e)‘ High Magadiw_ Beds and> Equivalentf ‘ BedsV I" of‘ the Natron\m ‘ v Basin The High Magadi Beds that outcrop on the eastern side of Lake Magadi have already been fully described (Baker, 1958, pp. 37-39) and little can be added to. that[YUM _L\4.L.IL.21account for ~ only a few exposures of these beds are found in the present area, in Eh;the 'MJCHM‘I'Hnorthern purlpart of,1 the J Rekereien valley. Equivalent beds connected with Lake Natron have,' however, notl‘t‘l been recognized previously, but as a result of the present work their presence is now established at the northern margin of the Lake Natron basin and in the area of the Ngare Nyiro swamp.

1¢.a-... Both these groups of beds represent the higher levels of Lakes Magadi.‘ andmi Natron,N;et1'n1L;1miand they were deposited after tectonic movements ceased in this part of the Rift111 Valley.Valley 111In the Magadi basin the beds were deposited up to a level of 40 feet above the present lake-level, and are represented by yellow-brown poorly bedded silts overlying laminated clays with fish remains. They pass below present lake-level and occur flooring the dry wui‘nr:southernn Limbarms of the lake, two of which extend into the northern part of the present area. 1.16F}

Two exposures of the High Magadi Beds131111. 11111."occur banked against11111111411 1111‘the lava111111 1'idg1‘ridge 1111on 1111‘the east side of the north Rekereien valley. In the more northerly 1111111‘some 25.25 11'feet-1 1'11of 1311115clays 111111and 11115silts without fossils are poorly exposed, 12 feet of laminated buff:1 111111.and 111'11\1.‘brown clays1111's resting1'1‘511'11g 011on 1101111deep brown unbedded silts. Similar silts outcrop to the south, and on both111.1111. these1. a: 1:11‘1.‘1.11':'1‘111‘Csoccurrences there111611: is a thin skin of saline efflorescences. A little further to the south,111111. 111in 1.111)the 1111ccentre 01111111of the Rekereien valley are low heaps of shingle with beds of fluviatile gravels which mark 111athe original southern shore of the High Magadi lake. The shingles and gravels. :1g1111‘131‘toappear to pass1.111.551. laterally into the pebble beds of the alluvial fan of the Orkajolasaien. :11 river.1.1101101151‘Close 11.1to the1111‘ western margin of the south-western lagoon of Lake Magadi,£11111:some 21200)1) 1111115yards 11111111north 111of the1111: present area there is one of the few places where the High Magadi1'11 BedsB: S are0 seemseen resting1‘=‘. ring 1111on older lake-beds. A small exposure in the banks of the stream that exposes 1111‘the bus'best ct11111section of Chert Series (Baker, 1958, p. 30, fig. 6, section1111.111 A)3111111111..shows the following section:-1111111: — 111111121inches Unconsolidated chert-kunkar-lava11111: ;_"1111"gravel...... (16 Kunkar limestone ...... (16

High Magadi Deep brown silts...... 241.11 .1. Beds {Brown clay with fish remains 4.1..

. Chert pebbles cemented by kunkar/ . . 2

Ch t S -11 er enes {Olive-green coarse sands. . . . ”124. The High Magadi Beds here rest with unconformity11'11'1‘“ 11111' on1'1'1 eroded1:1'111'101'1 and111111 1111‘;1111gdipping C'1‘11‘I'tChert SL.‘"11‘~.Series rocks.

At the foot of the north-west slopes of the Lenderut hills there are”11111“.outcrops'1'1118 of111‘ fluviatile11111 1111' -. beds at a level somewhat higher than the upper limit of the High Magadi11?.Beds..‘11ST1111.‘_\'They 21.1'1‘are exposed in shallow stream-courses near the junction of the Oldomut 111111111and Gelai(121115 motor11101111 tracks, where sections show brown silts and clays containing thin lenses“H1515 of1'11 lava1.111;. 11‘1'11_1§._r101‘_1—conglomerate. About one mile to the east a further section 20 feet thick of volcanic."_1111111111". pebble111‘. E111 11. beds11111111x'1151111with silt matrices is seen. In each case the beds are overlain by grey loessic 1111113soils 1‘11\'1covered‘1'1‘11 byl1.\' kunkar limestones. The beds dip at angles of up to one and a half degrees1~'1"".:11'1111.11111:~.111:'1northwards but the dips are probably depositional.

These beds are tentatively correlated with the High Magadi Beds and 111'1‘are regarded1111411111211 1.15as fluviatile deposits laid down at the base of the Lenderut hills during the11'11111111111111111411131011—maximum development of Lake Magadi. They are overlain locally by coarser‘1'11'91‘1'_'1‘1'.‘pebble‘11 beds11"1‘11 of'1“. more1.11‘1". recent.CL'C1'1 1.11113.date.

The beds deposited in the High Natron lake extend over' a11 111.111wide 11"‘1‘11area to1.1 1311‘the north11111111 111‘of Lake Natron and are seen on the Ngare Nyiro plain as1 far'111'11»1'1'.1'as the northern11111'11'11‘1'21 boundary111111211111} of(11‘ the1116 area. Exposures are poor as the beds have not been'. exposed1' 111111;..‘11 11'to any"11' depth11511111. by115- erosion,13111511111. 111111and in many places the surface of the plain is covered by friable11...‘ 111115clays 11111and wind-blown11.111114131111111accumula-111111111111 tions of light colour. On the plain immediately north and".1east of11'} the12111" 1.'11:1\121511outwash fan1115111111111)of the Pagasi river there are many low hummocks of fine grey or brown.111 loess11.13»: 11111111111:overlying 1:1)111p'1'1c1cdcompacted ._1~. brown silty clays showing poor bedding and having scattered efflorescences.111 113C: 01of1:13.115.salts. T111:The limit of exposures of these beds is only roughly marked on the map, for they' are1111‘ overlain11111111111 by113' more recent alluvial and loess deposits to the east.

The furthest extension of these beds is unknown, but it is likely that they underlie the whole of the Ngare Nyiro plain as far east as the Oloronga plateau, and probably‘rc.1=.1'11b1} extend1‘11'11‘111 for a mile or two north of the northern boundary of the area. The highest level11‘\-1‘1 reached1111111811 by11_ Lake Natron is not known precisely, but is likely to have been between 50' and.11 1001110112c1feet above its present level, as is suggested by the presence of poorly preserved11'11 beach111111111 levels1111115"i 1111on Melil hill ranging from the base of the hill to about 50 or 60 feet above present251.111 lake-level.1111\1‘—11"\"

The possibility that many of the other fault troughs. 1_‘1_11“1"'1111contain beds139111; 01-of High111_1'1'1 MagadiX1113111.11 age115.111 has been put forward in the preceding section. The‘ 5111112111115111111111111~rsediments occupying these111181 11'troughs1119.115 are1111: built up to a level surface and are clearly unaffected by_ faulting11111 01'or 11111:tilting.11.1. This11115 suggests that they are younger than the Olorgesailie Lake Beds,1:.1'11'11121which are111': affected1'. 11‘1‘11‘11 by111'111111111'11111115.minor faults. In any case it would be surprising if no beds contemporaneous1?:31‘113' 11111with the111: High1115.111 Magadi11112111111111 Beds were laid down in other troughs, for beds were clearly deposited1 ‘1'1A111‘1‘1 1111111“?under pluvial311111111101111111110115.conditions. -.

(3) SUPERFICIAL DEPOSITS The nature of the soils in the area is controlled largely by topography and drainage. The soils of the valley flats are usually light-coloured, fine and dusty when dry, and dark and clay-like when wet. They vary from light to dark grey, the latter being similar to the "black- cotton soils" of poorly drained areas. The latter are particularly common on the olivine basalts outcrops in the east of the area. In the Ngare Nyiro plain the soil is largely loess, sometimes redeposited by water. It is usually grey, structureless, and occurs in hummocks. On the lava horsts soils are generally thin and patchy, of red-brown colour and of the consistency of clay.

In complete contrast are the sandy soils of the Narsurana, Kisielib, Torosei and Olon- t' guru an valleys. Here the valleys have been covered by alluvial sands discharged by seasonal r torrents of the Torosei and Turoka river systems and derived from the predominantly quartzo-felspathic Basement System rocks to the east. Sandy soils also occur on the eluvial slope westwards from the base of the gneissic hills of Losirua,I« Lesiri' , and Lenkijabe. Pebble beds and boulder beds of fluviatile origin are found around all hills of moderate to steep relief, and a pediment of such beds completely surrounds Shornbole Mountain and 5 . , to a. . lesserLawn. U L\lLI;-.extent‘ .1r the!:| A Lenderut hills. The Pagasi7 alluvial. fan is notable:I- in that the greater I ' . . ‘ ,, , part'.Uiof T1:the material1 I'of whicha it, is kcomposed- is derived, k k" from a series, , of fans of kconsiderably, , greater'1' Ll‘lL‘,age.

. I '. . | . . '. , Kunkar limestone is: widespread'.iu,\l‘ LIL. in the area,I‘ and occursk.. in several, forms,. , the. A two| '.umost« important1 of which are "platy"D. kunkarx and, '"nodular"- kunkar.I, 1- The former'7 , is: not,- commoni ‘ ' and occurs principally in stream-beds and on old land-surfaces between the older series of lake beds. The "nodular" kunkar occurs either as nodules in soil or lying loose on the surface or as cemented nodules. The, latter, isI the‘ commonestL , ‘ 2 variety,1 andI L is,\. extensively\\.,. .".. developed on the Oloronga plateau, between‘L - Oloronga. . and7 .g Rekereien,’. north-westI. I; ‘of 01'li Doinyo Alasho, and more rarely on lava horsts such as Loloitikush.

The LUcemented nodular kunkar -of the i-Oloronga-RekereienV .area is‘ of interest:. , since it seemsackliix toLU be11' of composite age. On the ‘TL:Oloronga. . plateaui. ! itV forms. ' theﬂ; upper surface of the Oloronga Beds and slopes down westwards| ‘ andH passes‘ beneath,C ',|'the Ngare‘ if Nyiro“ V alluvium and the Natron equivalents of the High Magadi Beds. East of Oloronga, however, a similar kunkar overlies the Oloronga Beds and passes over the escarpments and descends into the Magadi-Rekereien trough and terminates a few feet above‘ the present, level- of. Lake Magadi.,1 - w This relationship is seen repeatedly on both' sides». ‘ of‘ Rekereien-the\" - , I I. kunkarH -L r liesg on I“an eroded surface of High Magadi Beds and on older formations. This episode of kunkar formation can be dated as early in the deflation. of the,, HighI ‘ Magadi' Beds. It. seems7 , that an, early kunkar formed on the Oloronga Beds, probably shortly after their deposition, and was instrumental in preserving them from erosion on the Olorongai plateau,' and' that7 there wasi‘- a second period of formation after the deposition of the High Magadi Beds. This younger kunkar has been eroded in the recent past,I‘. to form. * kunkar'- gravels, .,usually' mixed\,__ with chertH >7 fragments and lava. Superficial gravels occur as thin sheets resting on the younger kunkar limestone surface and on High Magadi Beds at the southern end of the south-western lagoon of Lake Magadi and along the base of the lava ridge on the east side of Rekereien. Similar gravels are often found resting on lake-beds in many of the small fault-troughs of the area, particularly at the north end of the Kibangaini depression. Wherever they occur they are found on the surfacenill'llNc or(13' mtoverlain by grey loess-like dusty soils. A typical sequence of superficial deposits .. in::1 the1h; centralscum part of the area is:- 4. Grey loess soil 3. Unconsolidated gravel j 2. Kunkar limestone 1. Lake-beds or river deposits

TheI he south-c:southernmostn" tip of the south-west lagoon of Lake Magadi and a group of feeble 9:1springsrims, occur at the northern boundary of the area. Both the lagoon and the springs (numbers -: 1.1 13 “Ink.and 14) have been fully described previously (Baker 1958, pp. 52-55). 18 V.-STRUCTURE The structure of the area closelyresemblesthat of the Magadi area to the north (Baker 1958,p. 61) and little can be added to the account given previously.

The metamorphic rocks in the south-east corner of the area are folded on axes plunging gently:CIIII )1'0to the1'.-1:101".north-north-east.I'IL . Three folds are recognized, the Ropet syncline, the Losirua anticlineIII1IIL‘1111: andand-1.1:the KileuK'. ‘Ic-I syncline.\I'I These folds are asymmetrical with gently dipping easterly limbs.lIn hs. TheT'I1'I: few131111121lineations10:13 and.1 minor folds that were observed are co-axial with the major foldstold; and11111.1 defineLICIInC IIa structural\ :I'IIL‘IIII' style with monoclinic symmetry. The lineations are therefore b-lineations. A1'. fewIQII lineations111L'I11 1.11% II!‘on Kaku hill on the eastern boundary of the area trend eastnorth-east, andIII‘I'Li may111:1} indicate.I‘IL a swing of the fold axes in that direction. These structures contrast with‘. those{1‘.-LN; deducedLEJLI by Joubert (1957, pp. 75-83) for the Namanga-Bissel area to the east,. and those11105: deducedIlL'LiIIC-I'Li by Weiss (1958) for the Turoka area.

The Losirua fault is inferred on stratigraphical grounds and was not seen in the field. On the basis of a tentative correlation between the stratigraphical sequences on either side of the fault, a downthrow of 5,000 feet to the east is inferred.

The remainder1II1111dCI' ofII? the area is dominated by fault tectonics associated with the Rift Valley. TheL‘ pattern:‘IIIIL'I'II is apparent' on Fig. 2, which shows the numerous parallel faults that characterizeI :/L' 111;the \II'.1L'_\'valley in11111the Magadi region. Little need be added to the account already givenLr-'.'-IL‘.:1{BI1'I.€I'(Baker 1958,WW pp.Il hi61-63)V with the exception of additional comment on the Nguruman fault1111111 552mm.system. The Nguruman fault, which first moved in mid-Tertiary times, was rejuvenated not onceIIIICC but13111 probably twice. A second movement occurred after the deposition of the KirikitiKirikili BasaltsB 1< and previous to the eruption of the bulk of the quartz trachytes. A third movement111II\C111L1‘-.I mustnIIis1 beII" postulated during the grid-faulting episode to allow for the cut-off of theIEIC OlorongaOlorLIII-jn Beds.BCd In any case post-quartz trachyte movements on the Nguruman fault at the same"‘31‘L' time1II'1: {15as [:‘Ik'the grid-faulting are strongly suggested by the fact that the quartz trachytes\ Lié‘dip towards:I.I'I\...rI1»11'c.the fault to the north and east of the Ngare Nyiro plain.

The pattern of grid-faulting111111;I isix well'- shown on Fig. 2. A feature of the structure is the repetitionI'CIICI111Ili of fault-blocks tilted1_|I.L11 to1.- the west and with prominent east-facing escarpments. TheseTE1C5C mirror:I‘I' the structure of‘ the" rift on the larger scale and strongly suggest that the mechanismn1;:Ci1.11I:\111 that originated rift movements continued to operate throughout the tectonic history111%1‘. .} Orofthe111: area.I-I LII. The Koora, Kordjya-Magadi and Ngare Nyiro-Natron troughs illustrate theihC largerIIIII'gCI' depressionsL!L.‘_1‘['L‘§5'IO[‘IS formed- ' by block-faulting and westerly tilting. In each case east-facing escarpmentsCDC“ occur with down-faulted westerly-tilted blocks to the east of them. In each (£155case the111C faults:II bounding these depressions to the west are slightly transverse to the trend of the131: majorityIII'I'II'IIIII} {Ifof smallern grid-faults, being orientated north-south or even west of north.

The larger number of grid-faults trend a little east of north, and form a structure consisting of narrow blocks showing step or horst and graben structure. Few individual faults are 1::11gc1'longer thanI11 1.. 1:11ten 1‘.-11.6%miles and. faultI systems are characterized by fractures arranged en echelon with rampsIamps betweenhCIwCCn IhLthe ends 111of III1_|adjacent faults. This is well shown in the Koora trough southsmuh of Olorgesailie.Olorgcsailic.

'I'IICIIIIIII-pl.111LsThe fault-planes 111C111

Ma‘s -- ,030'S

5,-7.0:020

1"‘I‘fhnw

2.9

22.3... 2

mm.»

all _.\\

2‘.H

,..

0 '-'

Fig.Mg. 3..2. - 'lThehe pzuwrpatternn M'of f'uulringfaulting inin the:in; Magadi\ldmdf 2‘region,C NJ”

-- - -- ,..

VI-GEOLOGICAL\I—GEOLOGIC'AL HISTORYHITTORY OF THISTHE MAGADI\I\ ‘1L_)I SISECTION‘5 TEDN OF 'I'I'EF.THE GREGORY RIFT VALLEY TI1L‘The southcr'msouthern part,11:1'1 11:“of LIIL“the (II'L‘g1'11'1Gregory Rift Valley in Kenya differs '1in scxcmlseveral important. own-1:11 ‘.'L‘sI1L‘L'Isrespects fromtrum II'IL‘the 111‘.1L‘"I1111‘tsother parts 11_-1to II‘Lthe north and south, but also forms the link‘x' I‘Cmc‘cnbetween II'I'L'the 1I.;mi111il111'dissimilar I1111but 1':..1ILL‘.Irelated T1119:1:1_\'5k.'1Tanganyika :'~L'L".1.‘1‘1'1section and the more widely known central‘1'11":1I 1.11111part I21in IIIL’the Naivasha-Nakuru111111:1SI1111——N.1 Ix‘L 1'11 'LII‘C'L.area. ‘1‘1I'1L‘1'L'11x1Whereas many:11_1 of the older rift volcanics and the structuresL'11.11"L‘.5z';associa-OCIL‘L— ted13d11I1I1II1-1-L‘111111'L‘I1I111ICIL1L'Iwith them are blanketed byI131 L"comparatively111"17 .I' recent volcanics in the areas to the10 11.111'north'1I1 and amuthsouth, I11II'1L‘in the \I;;u.1I1:1"‘L.11I1CMagadi area the g:greaterL'..11L"I' part of the volcanic succession is exposed. TI1'L‘The 111L-1211111'_11'I.1I1ILmetamorphic rocksI‘OCIL'S 1'1I‘1I‘11:of the BasementI? a'L‘"1'L'111 ‘System, which are believed to be"I’1‘ec;111.1I_1."I:111Precambrian I11in '1.age, 1111111form LIIBthe I"(\11111.I:11I1'.1I1foundation 1:11“.on ‘1‘1I1IL‘I1which 11:the 3'Rift Valley geology is built. TheJ'11151—post-CambrianC. 3-1.'1I‘11 :7'1'111 IIIhistory of tho.the L11‘L"'L1area Isis 1111-;one 111‘of I'L‘;1L'1.‘.1L".Irepeated uplift' and erosion, and erosion surfaces.1L‘CS OIof 011C-end- ('."1'L‘111c’L'L1119Cretaceous 11ndand SLIIJ—N'IIOL'sub-Miocene age_ have been. . recognized on the east shoulder11111.1:of the RITERift \'21IIL‘_\'Valley 1I11=.1I11:1'L(Joubert 1957,I957. 3.111.pp. ‘FJ—IIL‘I.9-10). TheII11.‘ win-'1"sub-Miocene peneplain is believed to have extended from111111 II'L‘LLIL'LIthe area 11111'11‘1north 11I'I\':I_II'.‘1.I1'.1of Kajiado 11911.1and Turoka,IL11'11I-L1.‘where it is well developed and overlain by- the111,: L1IL1‘stoldest 111‘:rift \CIIIILIIIICxvolcanics, 11L'1'1macross 1'11:the present111'L‘5L'111 LIday valley and to have been continuous with1 1111an L1'1'_1_I1'_111.erosion

FollowingI-OIIDMIIU, 1I1L'the CI‘LII1LIL111eruption L'1I'of the1:1L' L'111‘IILearliest of the Kirikiti Basalts, some of which'1 (11:131.:1'occur 0:1on the1I11'; upper1111111121 part3.11111 011‘of the Nguruman escarpment, a simple graben was formed 1.by}' apposedopposed normal11111111215 1'4111I1I11111'1:11I11;1\faulting on the Nguruman_ . . fault to the west and the Turoka fault to the east. The1:1. TIII'UWthrow (IIof theLI'IL" I‘111'11161'former \\.'1~.was 11111.1"1'1\;I1‘11approximately 2,000 feet and of the latter much less, perhaps between (100600 and.and 1,0001.000 feet,13:61. 8169111111111the throw of' the latter decreasing southwards till in the vicinity of Torosei no111' L'wIL'Icnccevidence of rift1'II‘1 faulting1111151111.; was.1111: I‘IT‘L‘I.“observed by the writer, although Joubert (1957) shows rift faults1‘2111i15 I11in II'IL‘the zIdIaL'L'n‘Iadjacent 1111.1,area.

TI11:1I:‘51\1.11c111IcThe first volcanic 911.3111;events I11in the1I1c \Magadi. area were the eruptions of the central'111r:.I \'r;1I-L'L1111'Jesvolcanoes ('.DIOI‘gL‘szIIIIL‘.Olorgesailie, 511011.11Shombole": Landand LenderutLender which probably pre-date the earliest rift-faulting.‘1'11'LII1I111. FollowingI_1'.1II1'_11.VI115_ILI16the buildingI111IILI1. a1of" these volcanoes there occurred the olivine basalt plateau vulcanicity which flooded the floor of the graben and covered the lower slopes of the central volcanoes. Contemporaneous vigorous erosion of the Nguruman escarpment is suggested by the occur- 1‘1.‘;1L‘-_trence L'II'of i111L‘:'I.1L“LIL‘._‘1LIinterbedded L‘1_11';|_-;IL1.1'.L:".conglomerates in the Kirikiti Basalts along the base of the escarpment. TII:The basaltL111911I1 e1'L.11,1LI1_1115eruptions were\IL'I'L‘ I'UII13‘1‘1L‘L'Ifollowed byI..1_ the first of the Plateau Trachyte flows (Lengitoto and Lengorale),Lenghralch \‘111IL‘I1which isIs 5131311seen 11I1Ig'only 11.111on the111LIxI1‘Kirikiti platform for, after the eruption of the first 11"11CI1_V11;'.~;.trachytes, '11a second 111'L1_I111'major 1111116111L111movement 1.1Iof 1‘11"the Nguruman fault took place, effecting a further 1LI0\\'111I11'0\\'downthrow OI‘of 1hr:the RII'IRift 1'I11..floor I1_\'by '.'1'.at least 2,500 feet. This rejuvenation of the Nguruman L‘S'L'm'pmemescarpment 1'63resulted.IIL'L‘LI in121 1I'-1:1'L11‘:the formation of a ledge of the Kirikiti Basalts on the face of the older 1'2112I1fault 69131escarpment.111.111.8211.

The PlateauFlam-.11 T.Trachyte:1L”.1}'L' '11'1.1‘1L'L111I1:11_11vulcanicity continued on the lowered Rift Valley floor, which'- I: 2.1at this time op-sloped1 gemLgently 111to the1‘1: we?west 6-since there was no rejuvenation of the TurokaIIzIL1-111 fault1:111I1011on [hethe east[toto compensateommm'te forI'1:"LI1r:>cL'1111the second 1large movement on the Nguruman fault. The trachytes are therefore‘1'C1I1ILIL'cstIn1I‘1ethickest in the west111ea‘111‘1i'.and thin to the east and probably never covered the whole width.\'1LIII1 (.1Iof ILI1L‘the RIII.Rift, .11I1311S1II1II1L‘at least in the southern part of the Magadi area. The trachyte vulcanicity '1\11.~1was broughtI1111L1gI11 to111 11a L‘:1'.1:"L‘close 1.1.I‘1I1with 'the eruption.,. of. a distinctive orthophyre-trachyte in the north, 11a 1:11.21lava 1111-11that seems 11.1to I111‘1.ehave 1.11'Eg1111.'11L'LIoriginated 1111111from 1:11.?the Suswa region, and will prove valuable when CL'H‘I'CIEIIIOI'Icorrelation OI1I1Cof the Magadi\‘IL‘.1_..11;II SIILL'emI1'1‘11succession \113‘I1with :11sequences to the north is attempted. The IIII'LIIfinal \UILL1111Lvolcanic 1.11I1115L'phase '1‘.'L-131I11':I1L11‘11was the formation of small scattered cones of scoriaceous lava 11:1dand 'L1=.I11ash. TI11.‘This activity111M} '.'1'111112'111'xI11I111‘11.r1.1_1:'1appears to have continued intermittently for some time, for the ashes 1.1I‘of .\I11:I1.11Alasho '11“1Lare L1'1I'.‘t'L'It111111'.111L"111.1wcontemporaneous '11I1I1with II‘cthe 111111"Oloronga lake-beds. 21

With the close of volcanic activity the floor of the Rift Valley can be visualized as a practically5.‘I';:-:.‘15<.‘:1H_\-' flati1. expanse of trachyte lava with the olivine basalts outcropping to the east. A.-\ number of small ash cones and the upper parts of the old central volcanoes rose above the level expanse of lava. On this surface the Chert Series was deposited as a comparatively thin layer of clays and silts, culminating in a fossiliferous limestone. These beds were then eroded and a thin kunkar limestone deposited locally and the Oloronga Beds laid down. These beds are poorly bedded silts and gravels and were formed largely by the redeposition of tuffs in shallow lakes and by streams. There then ensued an arid period of climate, the conglomerates of Pagasi were deposited as torrential deposits on the southern end of the Kirikiti platform and a thick kunkar limestone was formed on the Oloronga Beds. Following the deposition of the Chert Series and the Oloronga Beds came the grid- faulting, which broke the Rift floor into many parallel blocks and fault-troughs. This phase of faulting can only be likened to differential collapse of the Rift floor along sub-parallel fractures.:‘I'Llctt'L‘ It‘ was accompanied- ' by.' a second. . minor rejuvenation. of the. Ngurumanu‘rn: fault, for Elmthe ()ILOloronga Beds and conglomerates of Pagasi are truncated by it.' This grid-faulting L'Iepisode must have taken place quite rapidly, for there was virtually no erosion of the escarpments during the faulting, nor has there been much since. A third phase of lacustrine deposition occurred subsequent to the grid-faulting-these are represented by the Olorgesailie Lake Beds, which have been studied in detail on account of the vertebrate fauna and artefacts found in them. These beds are approximately 175 feet thick and rest on an eroded surface of volcanics. They represent a period of lacustrine deposition during which there were many recessions of the lake margin and during which Acheulian man camped and hunted along the lake shore. Reddening of part of Bed L8 (Baker, 1958, p. 34) suggests that at one time at least the emergence of the bed was of sufficient duration to allow the formation of a soil profile. Deposition in the Legemunge basin was almost certainly brought to an end by a phase of minor faulting which affects the lake-beds and caused local rejuvenations of the grid-faults. The 01 Tepesi depression was formed at this time and deepening of the Koora trough to provide an outlet for the lake is suggested by the presence of hanging valleys on the east side of the Shanamu escarpment. The 01 Keju Nero, which previously had been the main affluent of the lake which formerly occupied the Legemunge basin, then cut across the lake beds at the base of the northern slope of Olorgesailie mountain. A series of coarse boulder beds were deposited on terraces at an early stage in the erosion of the 01 Keju Nero valley. With the rejuvenation of the 01 Keju Nero a considerable amount of brown alluvial, and some fluviatile, material was laid down in the northern part of the Koora trough. Following the period of minor faulting a phase of lacustrine sedimentation took place in the Magadi trough. The High Magadi Beds were laid down on an eroded surface of older lacustrine beds and lavas. Kunkar limestone is almost ubiquitous on the land surface beneath these beds and is suggestive of a period of aridity preceding the onset of lacustrine sedimentation. The High Magadi Beds were built up to a level 40 feet above present lake-11h?— level, where a well marked strand-line exists with a talus accumulation above. Equivalent beds were deposited in the Natron basin up to approximately 60 feet above present lake-level. A lake also existed in the Kibangaini depression south of Magadi, but unlike lakes Magadi and Natron, it has dried up completely. Much of the sediment which clearly fills the multitude of fault depressions in the region was also deposited during this phase; the few poor exposures indicate that the smaller depressions were little more than swamps and are filled by alluvium and fluviatile material of various kinds. Wind deflation must be invoked to account for the partial excavation of the Magadi trough to allow the deposition of the Evaporite Series, which is probably stratigraphically continuous with the High Magadi Beds in the centre of the trough. The deposition of the Evaporite Series is connected with the onset of alkaline spring activity and is not necessarily due to a change in climate, for no positive evidence of such a change is found outside the Magadi trough. The interbedded clays of the evaporite sequence were almost certainly derived from the erosion of the High Magadi Beds and the 't building up of the series to the present-day level of the lake is largely the result of crystalliza- tion of trona and the deposition of clays in alkaline lagoons in comparatively small depressions on the High Magadi Beds. The deflation of the latter would very likely be assisted by the fact that the alkaline waters would inhibit the growth of vegetation on the lake flats. During the period of deflation coarse poorly sorted boulder beds were deposited at various levels at the margins of the Magadi trough. The gravel sheets spread over the surface of or"

22‘N\

111Vthe 1.11\1.f1.11~lake flats .V‘Vat 111Vthe southernM111 1:".1 V1end 111of 1.1.1.1:Lake \1.‘1g.1.11Magadi V111V1and 1.1111the 51:10}grey \11'.'_\silty .110xx—111V‘.loess-like 11003111111111111111accumulations in the{311‘13lee >11“side1 V‘ 111111of the trough suggestC51 :11111that 1311:.21‘1V101111111110115the arid conditions \‘111'1011which 011113011111011111101‘111gcaused the lowering 011110of the High Magadi lake1.1.1113 andV11: 1' the erosion1.1111. 11‘of its11~ 4:11.115deposits 11.1.0have persisted1101-1151011 5'V11V151.':1‘1i;a1‘;1substantially unchanged11:10111111g011 to the present day. VII-PLEISTOCENE(CHRONOLOGYHR1)\'()1.()G\' AND\\'D CLIMATE As a result1 of the work.1 . of Leakey,1'10.\.I~.V'_1.. Wayland,\\'V13.1V1:111. \Nilsson1\\1 11 .1111and 111:101‘~'others on 1110the Pleistocene1510151000310 deposits in EastE111 Africa.\1'1'10..1 aV1 faunal1'V1V11‘V ‘1 1.11V1and culturalV‘11.111:‘..‘1 \V‘V1'V11‘111'V‘sequence has1‘ .1» been110011 evolved011111011 V11111and used113011 to111 111110date various deposits.' On the' evidence' of the1111‘: lithology.VV‘.1V~1.1111 of1111' various1V1“ 11.1% fluviatile,1111\11111111. lacustrine111015111110 and $111sub-— aerial deposits and‘V1VLV1 their1.1V contained. fauna. .1 aV1 011111V111Vclimatic \V‘V111V1VV‘sequence 11:11for 1310the PleistoceneP101511100110 11115has 111511also been worked out, and1‘.V'1‘. a seriesV‘x of111'1'111:1“.11.V1‘.four pluvials7V1:111and 01911611corresponding11111 :111101‘1111'VV1111151.111(‘11111‘1'interpluvials of varying length and character have been1110011 postulated.11V1>1V11.111:V1. Terms11511111111010that were 111101114111originally 111011115011proposed 111rfor pluvial1111111111 phases have now achieved stratigraphical1111111031 .1significanceV;111V‘V‘ 1:1.11111102.(cf. Zeuner, 1952,1‘15: pp.1111. 3-1")249-252)211111111and this tendency has recently culminated.V1VV'V1 11‘.in 11:10the VVV‘00111V121V‘Vacceptance 111'of 1110the 1011115terms Kageran,1x11g011111. Kamasian,1x11111as1'V121V Kanjeran and Gamblian as stratigraphical' 1‘V111111V‘V11 1111113units 13111(3rd Pan-AfricanP:111 \1'1‘10V111 (111131055Congress 1111on Prehistory,1’1‘011151111‘1'. 1955, pp. 31-32). These terms refer' to the:1" V1;:111~:1'\deposits laid11111 1111.11down 1111111111111111111111‘101211111115101111‘1‘during a pluvial period together with the succeeding interpluvial period1-V'11 V111V.and 111‘1'are V.V"defined11116111111.(op. cit.). .1115as '51"stratigraphic“V1[1"1‘VL'.‘\111C 111111climaticL1111" divisions". A stratigraphical :1111110110'.nomenclature11110 111SV‘11based 1111on ..a 111:.1111'11011V‘V11hypothetical 0.1climatic111111110 50111101100sequence 15is 11:of necessity imprecise and controversial,'h1111011especially when 1111'the geology1111.101 1111110of the 11110type 111011111109localities 1111on which it is based has not been adequately described,1‘11VV1 V1111and whenV11 1111111what 1\is 1\:1111111known 111'of 1110\0these 3.10%areas can often give rise to conflicting climatic11‘.1.11 V interpretations.11::V1;11 1V111111‘1x. In111 511110spite of{11111.0(-these CL'Ilix1dC‘1'L1111‘1'ziconsiderations suggestionss11;7‘11“ 111111»- of111' :1a 01:101.glacio-pluvial1.111 .111.:1 V‘1;‘correlation1‘01V11 1.111 111111with EuropeL111‘11110 111110have 110011been 11111110made 1‘C'1‘110.1‘~)54.(Cole, 1954, 1111.pp. 51-55),U and correlations.C1V11'.i1.'11‘1\ based111.10.91.11on climaticV‘1111.11V‘11V1V1111‘111deductions..11:110have been1100:11‘0g10;11011111111110repeatedly made in111110111151.the past. Cooke("111110 (1955,111/55. p.;‘1. 51)511 has'f1‘.‘ recently drawn 1:110:1111111attention to$1.1 these:11‘11.1':.'Vis to be of value. The evidence3V11‘111V1V“ from.'1‘11111 the11111\Magadi..:. 1111. region1V 1111.111. supports71111111” the111V 01111001111111110concept of the C1V1z111.111111Gamblian 1111111111.pluvial, V12111and an indifferent case can’ be made' for' ‘ a Kanjeran' 11111 pluvial,1 11. .1.11 but111.'V the1110 011110.1001r111111evidence in favour111 111of the1110 latter is inconclusive. The Olorgesailie Lake BedsCV.\ VV1111‘V‘1‘.'1are the only firmly 111110111dated 1111:1051»deposits in'11] 1110the \1115Magadi’V11i1 region, and are regarded as comparable'11‘ to.1.) Bed151311 IV1\ of the C)1C1‘11\'\‘V11Olduway SC‘QVICZsequence115C 111113and to the Kanjera Beds by Leakey (1952, p. 209; 1951,1‘151. pp.11:1 29/31129-30; Kent,1.1 1942,1042. pp.111.1. 125-127).135713“ These beds are also correlated with VV1.diatomaceous1.11111 1VV'11. ..1110-1V11\lake-beds V11at Kariandusi1\‘ 1111.14 (11111and Munyu\‘1111111; wa Gicheru (Shackleton 1955, pp. 259-260).211111. These1111\0 beds.11:i 1111‘are upper111‘111'1‘ 1:11V1V111‘middle Pleistocene1161>11‘C1‘1‘.C 1:1in 1.?age.. The succeeding High Magadi BedsR01 .111and1111.11:' their NatronV\'V.11‘V‘11‘. 011111111011“equivalents V110are not1111: positively111151111011 11111011dated but can reasonably be compared with ‘the former" ‘ greater7'C111L'1' V'\t01‘.\111‘1.\extensions 1.1"of 1Lakes.11;1‘\ Naivasha,.\ {~1‘ Elmenteita and Nakuru which have been referred to the0(}.11111711V1:1VVGamblian climate‘ phase,1‘ 111\L'V which is regarded as being upper Pleistocene in age (Cole 1952,. 1111.pp. 4‘)49-50,511. (11.1.1.60). The Chert Series and Oloronga Beds are therefore earlier11' 111.111than 1.1111101upper 11111010middle Pleistocene1’101>:11V‘1“1:1‘ in age and are tentatively placed in the lower Pleistocene. No.\1:1 01111011111011correlation 111111with 1111101other 101101"lower Pleistocene beds is possible because of the lack of fauna111 1.11‘or \111110stone tool111111 0111111105cultures in111 1110111them 111111and no climatic evidence of regional significance can be deduced from them. The High Magadi Beds and their Natron equivalents1 1 011% must111.151 be110 regarded1‘1‘111V‘12‘V101‘1 'V-Vsas having11111'1ng'been11.110: 1 deposited during a period of appreciably higher111~:‘V11V"V rainfall1.11.1111111111111111111i1‘1111111gV11111'050111.111111than that obtaining at present, and so accumulated under pluvial conditions in 1110the «1150sense 1.11of \\Wayland. 111111 11953.(1952, 1.111.pp. 5‘)59-61).{11].T11V'The evidence for dry conditions before and after the deposition" ‘21 1‘111.1.‘)"1‘of these beds1321311;has\ :V.already been mentioned. The evidence for pluvial conditions during the formation‘1"1111‘. 1.11. .11: 111of [:10the 01111»Olorgesailie51.1 Lake Beds is poor, for deposition may well have been brought110V11111V11‘.to an end1‘111‘. by11.; tectonic1V0111111V‘ 111011115.means. In addition, evidence for dry conditions immediately previous1: V1.1111< 21.1to the1510 1101111511111deposition 111'of 111V-the Olorgesailie Lake Beds is scant. It is notable, however,‘. that111.11 in 1111\0Viclosed 111Vx111sf11‘1110basins in the Magadi\11101101 area which have a similar form to the Olorgesailie lake1.1110 basin11.1411: 11ndand 111111011which have111110 .1a 51111i1'V11'similar I rainfall no lakes exist under present-day semi-arid.V1 conditions.01111111110111 This11115 51;gg0<15111111suggests that 11101111111111}the rainfall may have been higher in Olorgesailie Lake Beds times11.1101 01111111111;-than at the 11050111present V1V1g..11111day, but 31110-01110since the existence of the Olorgesailie lake was to a large“10 degree1101110: dependent110 .1011V10:11 1.111on 1110the 1'111‘11111110nformation 111'of suitable basin by tectonic means i.e. the grid-faulting,1111110. 111111then the1110 11111111111111initiation 1111111110and the ending 111'of lake deposition may well have been controlled largely by11;. tectonic100131111“ movements.11111101110111). The1‘;- 1:.1VZV*1‘VV‘V‘evidence for11V climatic change before and after“1120the depositionVHAV11V.\‘111‘111.11.'111V‘of the HighH1211 Magadi\1151'V‘112-i Beds130115 and:11111 their11101101.1111‘1..‘11V‘.11\‘equivalents is: good,1.1111V1VV1.111and that11111 pluvial111111111100111121111115conditions existedC‘\‘1\11011 in111 Gamblian(31111111111111 times.[1111115 The evidence for climatic changes previous“10.1.11.“ to11.1 these1.1310 in the1.10 Magadi\1‘.1f_‘1.1111 area11103 .110.are, 11111101111however, inconclusive. 1

VIII-MINERAL DEPOSITS An exclusive prospecting licence (E.P.L. No. 103) is held by the Power Securities Cor- porationi" ilLi‘a Limited,._-. . Nairobi,.‘ul over the whole of the area south of Magadi, excluding the areas underlain byl}_\-' Basement”mum- System rocks. The southern part of the area leased by the Magadi SoduSoda Comp-anyCompany Limited,111111tfalls into the present area, and includes the land bounded by latitudes 2°I (HY00' and.and I2° 0505' .‘5S., Landand longitudes. 36° 10' E., and 36° 20' E. Under the terms of the lease the[I'm Company(\:n‘.|"1=.|‘.)'.has' the_‘ rightst" ‘ to work and dispose of deposits of sodium carbonate and other “I -. . 3.2.5.11;salts mixed with it, but has no rights over any other mineral (cf. Baker, 1958, p. 64).

1. Kyanite KyaniteK‘JJHiiC gneissesgncjsgcs outcrop prominently among the hills in the south-east of the area and occur ini. three‘IH'cc main:1 ' beds, along the western slopes of the Kileu ridge, on the northern and \‘alL‘:western slopes of Losirua mountain, and on the ridge east of Ropet. In each case the rocks are biotite-kyanite gneisses with variable quantities of muscovite, quartz and felspar. Only the Losirua kyanite beds have been studied in detail from an economic point of view, but the similarity between the three occurrences suggests that the conclusions reached for the Losirua occurrence can be applied in a general way to the others. The Losirua kyanite bed is approximately 300-400 feet thick. Only the lower 50 or 60 feet contain appreciable quantities of kyanite; the upper part consisting of thin kyanite- bearing layers in biotite and biotite-muscovite gneisses. The lower kyanite-rich bed is homogeneous and contains greenish kyanite prisms up to half an inch in length. A series of grab samples were taken at various localities on this bed by A. L. Stewart, Government Metallurgist,f\-'1c1al|=\i1'gim_ and the proportion of kyanite estimated by means of separations in heavy liquidsElqmds andand. by electro-magnet. The localities are marked on the sketch-map, Fig. 3, which also;E.\'0 ‘LHC-"ARIshows the1' geology and form of the deposit. The percentages of kyanite estimated by Mr. Stewart are as follows:- Sample No. Kyanite % 220/1 12.4 220/2 9.8 220/3 18.6 220/4 18.4 220/5 12.8 The-. lowest:L'1'-.‘.CSL 120‘15 feet of the kyanite gneiss at the locality of grab sample 220/4 appears to be the best part so far located. The same horizon sampled at 220/2 gives a low kyanite percentage in spite of promising indications in the field.

Three further grab samples were taken, in the Kileu band of kyanite gneisses immediately east of the closure of the synclinal structure, three and a quarter miles east-north-east of the summit of Losirua. Kyanite percentages were determined by Mr. Stewart as follows:- Sample No. Kyanite ~ I %q 220/62:“ 9,7‘ _ 22"! " 10,4.. 1 220/7 «a. . . I 220/8-v-—|-: 8-4I

The main-:Jriiunproportion 0'1“of muscovite in these gneisses is generally higher than in those on Losirua.Lemma. More comprehensivecomm sampling of the Kileu kyanite beds is required, however, before an1:1".- -:>pfni::nopinion ofn. their economic possibilities can be expressed.

j 24:4

\- I I L f 7' 1 l ' \

r: .:..: ".:.~ Quartzites 1(((/1\ / MuscovIte-biotitebIotIte-garnet gneisses5ch'st and Muscovite gneisses I~~~ 220/2 ,.. Samplelocalitiesandnumbers illIIIJ]Biotite-garnet gneisses Stnke and dIp of foliation I I ~ ~I B,otlte-k yamte gneisses Scale 0 1 2 t-l,iLES J

Fig. 3.-Geological sketch-map of the Losirua kyanite deposit" The kyanite gneiss occurrences on the ridge east of Ropet were hastily examinedCU. andL'Izzi it is clear that only the western bed is likely to be of economic interest. TheJ . '_bed\_'. isnil-11:110-homo- gen~ous, moderately coarse-grained and very similar to the lower part of Li'icthe LosiruaLusil'uu kyanite3:»;12’liIL.‘ gneiss. These figures quoted above indicate that the kyanite gneisses are less rich than might be suggested by field estimations and that the deposits would be a sub-marginal economic proposition under present economic circumstances. The economic potentialities ofL)!‘ the[hi3 deposits are also adversely affected by the distance to the nearest railway, which is approxi-wroxl— mately 40 miles, and the absence of a natural water-supply. The deposits are, however, large, and should there be an improvement in the price and demand for kyanite and mullite it may well be profitable to carry out a full-scale investigation on them. J

25'1

3.2. (21111111111:Graphite 5511.11.63Schists Kyanite-graphiteKVaITFC— ‘__1.r:.1 1‘|111L' 1611\1x'schists 1.166111”occur 11'near the top of the quartzite at the base of the kyanite $111L‘1S5gneiss L111on Losirua.1051 Few1-611 6x1111~11'56exposures are seen and no estimate of the thickness of the graphitic bed1161.16LL1L1.can be1'16111:1<1'L‘.'1'1L‘1made. The graphite111311; occurs in flakes up to 5 millimetres in diameter and locally I 1'01'2113forms 1113111111111perhaps 21120 1161'per 66111cent 1'11'of L116the 1'-.‘rock by volume. j '1'111.~'This 1166111161166.occurrence, 1115‘1‘161.however, is not likely to be economic, but could be investigated by pitting1.111111. 1.: 5111111111161611611.should there be a 11311111.:revival 11:"1of interest in the neighbouring kyanite deposits. J 3. Water Supply '1'1161'6There 'L11'L‘are \‘j‘1‘L‘1'.11several perennial116161131111 1x1.sources of water in the South Magadi area, and in this respect16<11661111\.~‘11"1111"it is somewhatL16.'1161'<111_111'better supplied than the area to the north. The Ngare Nyiro and the 5.111111111101111swamp north of1'11L'11‘11‘111'L'1.1'Shombole provide1111111116 ample water and also good grazing to the many herds of Masai.'\1'.15111' 6111116-cattle 131:1:that 6".crowd there in the dry season. The lower reaches and the out-wash fan of111' theS116 Pagasi1’11111155 river1'11 61' alsoL provide water, except in very dry seasons, but this area is apparently avoided;1\'L111 by the Masai, possibly because some of the forest there contains tsetse fly. The (11.11131other major watering place is Lake Kabongo, which has been perennial since the unusually heavy1'16'L1\ 1. rains in 1951, but which the Masai state dried up seasonably before that date. This water is normally extremely foul because of the large numbers of cattle who crowd into it in the'16 dry111'} season.566.5011. As many as ten or twelve herds can be seen watering at one time in the dry' ' months1111.111111< while\11'11'1‘11‘ others are struggling over the escarpments that bound this lake on three sides." A\1'111'1'161'further supply is obtained from wells dug in the Torosei river near its junction with the1111: SinyaS::1_V'.1.!Landari.1:111:11'1. A large number of wells are dug here each year into the sandy river-bed, which‘1'111611 has been1.16611 partly dammed by a low concrete wall dug at the point where the river enters a1‘. gorge.gL-L'gc. These17161.16 wells1 are likely to be perennial except in unusually dry years. SeasonalS-L1:<111"" water-supply\VL1L61'-.~'.'.1111'1. is1- obtained from a number' of small, undeveloped natural pans, from1'r'111‘1 rock1'11611' basins1‘ FM in31'. river:' 1161' beds,'1. and from wells in the bed of the1'1- Turoka river. Pools of water occur1166111' in1:1 1116the $611.1?11L"‘1':southern part.1L‘1"‘ of the Kiambu valley after151' rain, and‘ also along impersistent 51111111111.shallow \V‘E1‘.‘L"1'L‘1'.‘111'\'11_’\'watercourses 1111on 1.311.“the M'balbal Ormeresho plain. '1'11'6The 111mmovementsCI‘W‘ZHH' 1111'of 1111‘the Masai are controlled entirely by the necessity for finding grazing .‘11‘L1V1.'11.61'.and water, 11:111and 111.11'1‘1during '111:the 1.dry season they concentrate around the few reliable water sources, \V'1t11with resulting1‘CSL11111‘11' .961serious1111.11 1.116overgrazing of the surrounding country. The Losirua and Kileu hills, 1'11“.on \\'111611which 1.1161there1'6 1:is 151.1061good grazing, are neglected on account of the lack of water, and the northern1'1111'111-'61'11 part11211'1111'1.of the M'balbal\1'1.1. Ormeresho is undergrazed for the same reason. Since future L16V61'L111111'development11111in the. area is' likely to be in the form of grazing control, a planned augmentation 1'11'of the1116 miningexisting water-supplies11. will be necessary also. The problem is to keep the Masai away from1':'1.11'1.1 1116the potentially11016316611} better1761.161 grazing areas, such as the Ngare Nyiro plain, the ground around ()101 DoinyoD11111y11 Alasho1 1 {13111.1 11111.and the larger valleys such as Lesoit, Oldomut and Narsurana, until :61”grass1s 0.111can re-establish1'6——L51.'L111.11\11 itself115.. in these areas. This can only be done by control and the provision”111.111.511.11 ofK‘ additionalL1L1C1'1101211111L11L‘1'—.\1_1‘,11111L‘\'.water-supplies. Initially1111111111} the1116 quickest1.11116 and cheapest methods of increasing water-points would be the 6.1111311111311011construction 1'11'of "1.11.110""tanks" L1or artificial. deep pans in suitably chosen spots. The best sites are those1110562111‘cady0661.111113L1already occupied by1111 natural:11. pans and by places receiving appreciable run off and which have11:1\'6 reasonably‘1"L‘""50Dl1131" impervious11111?Ci'\'1“1 15 E111.soil 1'1.'of adequate depth. Many of the valleys of the 6.1:central111 11:111.part 11'of 1'16theLareaarea L-11'6are 11L‘1{_‘.1"L‘11floored 111111with 61:1)?clay soils of considerable depth and transient streamsM11611flow 1111.11into them.111C.131.Tat11(8Tanks CL1L11L1could 11-6be constructed1L'1111x'11' 1161611 at the points where these streams disappear,11113.1'. :1111.and 1:01:16.could 116be 56.so 11rr11n16darranged 111.11that L116the ststreams1215 flow1.‘ 1.1'1 into111.11 them.1. Such tanks would, however,61'. '61:be S11111'1—.short-lived j 1111on 2166011111account 61of sediment3'L‘L111T1‘L'1111'.‘carried1111LL1 1:1111into them.

T111316There 11.16are many"113.111"suitable.'.11111.11 1; «11.65sites for11'11 tanksLL'111ka1'11on the IlkoritoritiL ..and Ormeresho plains, on the j 1'11'0\H‘l C111_V5L111 1111111111 1"Cg1'L111. 511111111‘11111111511111.1'6L15'L11‘6'L'1'1 I brown clay soil of that region. Similar promising areas are at the southern end of Oldomut, 111;at 1116the 50111116111southern 61111end 111'1116of the 561'.Sereata valley, 111at 1.11.the northern11161'1‘. 6:161.end 611116.of Rekereien61'6161111111111-1and in 1116the Lesoit65111.. ~ valley.

No bore—110165bore-holes have11a16 been116 drilled11'1111611 111.in 111Lthe"- :11'12‘1'1.area, 1'111‘1.but 11116one 11111161.drilled“‘ 1L1a few2"61.‘ 1111165miles 11011north-east11-61151 01'of Ilkoritoriti11kL11'1tL1riti proved to be1.16 L111}."'111Cdry. The L11.'1.11.'6.

l 26

4. Economic Possibilities in the Area The only mineral likely to be of economic interest is kyanite, which has been discussed above. Future developments in the area will probably be in the direction, (‘21 ofLL improving. water-supply, implementing grazing control and in the provision of betterascr' motor-tracks.mmm—zmuks. The area south of Magadi is likely to remain very much as it is at present'2‘. for1m {1:121}:many M7153years to come.

IX-REFERENCESv*\' N. \. ~ . m x 1 ; ‘ 1 ’ ' ‘ ‘ ' ‘ l ‘ , . .1 \ 1 ‘1 ‘ .L1.ﬂ\1\1f',. U Bain, G. W. and J. H. Beebe, ‘V.1954.-"Scale_‘ _‘: _VModel Reproduction of Tension Faults". Amer. Journ. of Science, vol. 252, pp. 745-754.

Baker,)1 V B.1 H.,11, 1958.-"Geology_; W : of the Magadi: Area".~ Rep.K. No.xx 42,1‘1- Geol.\‘y v1.5.11.Surv. Kenya.“Htx‘ 11 Cole, S., 1954.-"The Prehistory of East Africa". Penguin Books. Cooke, H. B. S., 1957.-"The Problem of Quaternary Glacio-Pluvial Correlation in East and Southern Africa". Proc. of 3rd Pan-African Congress on Prehistory, 1955, pp. 51-55. Dixey, F., 1956.-"The East African Rift System." Co!. Ceo!. Min. Res., Bulletin Supplement No.1. Gregory, J. W., 192I.-"Rift Valleys and Geology of East Africa". Joubert, P., 1957.-"Geology. ‘1 of1 the111. Namanga-Bissel\ v.13}: 1w Area".‘x, 1 Rep.[hp No.\v 39,“1 lmGeol,1 Surv.SUN. Kenya. Kent,ix». 1, P. E., 1941.-"The,1 11Recent., History‘ , and1 Pleistocene’ “,1 , ‘1 Deposits:‘u ‘ w 1of the.1 Plateau)‘11'L111. north1 of'1 LakeEL , Eyasi, Tanganyika".I 1~ Ceo!.1’“ Mag., vol. LXXVIII,k\'\‘l'1 pp. 173-184.'37'm

iiiii 1942.-"The,1: ‘ Pleistocene'1» 1. Beds_ of Kanam‘ and‘ Kanjera,K1111“ Kavirondo,f‘\ Kenya". Ceo!. Mag.,'11" , vol. LXXIX,\> 1 pp. 117-132. , ,i if 1944.-"The‘ J 1 Age, and‘ Tectonic7 ' Relationships1‘1. ‘ of Eastr African‘x w . \w-icazVolcanicW Rocks".{:1 1; 1 »Ceo!. Mag.,, vol.1 LXXXI,‘ pp.,1 15-27.1 1‘ Leakey,L111 L.1 S.~ B.,1 1951.-"Olduvaif , K Gorge".‘1 : , if 7 1952.-"The" ‘ Olorgesailie1 , Prehistoric‘1 K Site".\ Proc.1‘11 of the" 1st1»‘ Pan-AfricanP117 Congr.(7. 1111’don Prehistory,. 1947,1 f p. 209.i‘ Parkinson,.1 LIV ,1,J., 1914.-"The1L1 '[v East1 African‘~ troughg in1’ the1*neighbourhood: ‘1 1 ‘1 of:1"the Soda Lakes".IL1«.;~ (JCeogr., Journ.,1 vol. XLIV,Q ‘ pp. 33--49.3 $3Schoeman, ‘1J. J.,1 1949.-"Geology‘ A , of‘ the Sotik, ‘ District".1‘, Rep.\ No.‘ w ]16, ilow‘.Geol. Surv. Kenya. Shackleton, R. M., 1945.-"Geology of the Nyeri Area". Rep. No. 12, Geol. Surv. Kenya. 1946.-"Geology of the Country between Nanyuki and Maralal". Rep. No. 11, Geol. Surv. Kenya. 1951.-"A Contribution to the Geology of 1the Kavirondoixwmlxyzhiw RiftR131 Valley".\ilcjfﬂ Quart. Journ. Ceo!. Soc., vol. CVI, pp. 345-392. 1955.-"Pleistocene Movements in the Gregorylt':\'.1i1u}”.Rift Valley". (InCeo!. Rundschau,,Rsmfwcixw. Bund13a AZ}.43, Heft 1, pp. 257-263. Smith,51711111 G.(73‘ E.,hf, 1907.-"From‘1"‘1,7_,~"‘r‘,x "1 theI“: Victoria“x .3“ Nyanza.\> 11 to1:.("mKilimanjaro". Ceog.(iwcr. Journ.,./1.r:;.*'2:.. vol.\ :11. 29,1‘). pp.1‘112497273.249-272. Smith,11111. W. Campbell,(I , 1931.-"A classification of some rhyolites,(x. trachytes1r;;;h_\'rca and(my. phonolites,pfmnolil 5. from:‘m'n partpgu of Kenya Colony, with a note on some of thehe .,.\$1‘1;1111L‘dassociated bdx‘niti;basaltic rocks".mckx‘I Quart.(:7; Journ. Ceo!. Soc., vol. 87, pp. 212-258. (Mm Uhlig,l 1: ' c.,(7.. 1909.-"Die1")1’1‘9: ”2%)1, Ostafrikanishe Bruchstufe und die angrenzenden,_ . 113:1? Gebiete Zwischen7" “H denmm SeenSum Magad\1 ‘ggd und.: Lawa ja Mweri sowie den WestfussLI‘M desdc< Meru,\1c1'11. Wissenschaftliche\\'i\\‘cmc} 211'!“t 1 1‘ * . 1 Ergebnisse" 1m. der. .1 Ostafrikanischen Expedition der Otto Winter-Stiftung.\\ :mc: “€11 3111; TeilIt}. 1,f Dit';D7: Karte, Mitt, aus den Deutschen Schutzgebeiten. £rgang.':__' No.2.\H ,

IL 27

Uhlig,L'Mig. C. "andand 1-.F. Jacgm'.Jaeger, 1942.-"Diel‘ML— ‘L)1c (")Ostafrikanishefiu’zwlxnrmh Bruchstufe und die angrenzenden'I {C 3‘. SCH (“whimsGebiete Z\\'ischcnZwischen den ScanSeen MagadMngnd undm6. LawaL“ LL LLja Mweri sowie den West Schlussc.1115: Ger;des I .\-1eru".Meru". \K'issunsghaWissenschaftlichefihdm Veroffentlichungen‘v'en'efl'm{hdw-gcvw Ms.des Deutschen Institutes .Lfur2' Lander-i..;t:n‘.-;:'— I kundc.kunde, NCUCNeue FolgeFoigc H}JO. WulmcrWalther, P., J922.-"Sodium1921—”Sadzmn {fui'bomtcCarbonate Minerals\Imcxtm wmfrom the.m.‘ MagadiV-Eugvni' LJ-J‘u.Lakes, British East j A:Africa". rhm’f‘.Amer. Min.,UL”. \03.vol. 7.7, DZ".pp. ‘93—‘51.86-88. Wayland,\‘x'nylund‘ E. J..J., 1952.-"The1952‘ --"Thc Stud}Study of PastPmL (Mung-aClimates in11'; "l*cw'.‘.11Tropical Africa"..1\.‘_‘:.'i,_'_L". Proc.Pt'nu. of 1st15: Pan-AfricanPun-African C'OHgI'.Congr. on Prehistory,{’I'chiirol‘}: pp.m“ 5959-65.(-‘5. “bias. I... E‘. 193. .— "Summural ‘\I‘.C;[_\Si.\‘t."1“‘1r:Bsm‘iTJMSiC‘l'JLL ILL'URL‘. Rwy; Puff 'L' Weiss, L. E., 1958.-"Structural Analysis of the Basement System at Turoka, Kenya, Part 2". f ("(u’.Co!. Owl.Geol. Min.UM. Res.,Rm". H‘l.vol. 7".7, pp. 123-153.1:34:35. Zeuncr,Zeuner, F. E.E., 1957.,-1952.-"Dating"[3HLingihCP.151”the Past".

, I

(111K.Q.P.K~484-1m-~/614515 1m" 2 M

... -