A PRELIMINARY SURVEY OF THE POST-GLACIAL VEGETATIONAL HISTORY OF THE KASHMIR VALLEY
GURDIP SINGH Birbnl Sahni Institute of Palaeobotany, LucknolV
ABSTRACT course of the main river Jhelum. The latter, along with its large number of tributaries, Pollen-analytical investigations of Post-glacial has shaped the presen t relief of this area, deposits from four sites (Toshmaidan, Braman, Walanwar and Damamsar) distributed over the resulting in the extensive dissection of the Kashmir Valley slopes of the Pirpanjal Range Karewa beds and causing the formation of constitute the subject matter of this paper. The terraces and lowland swampy areas and study reveals for the first time the Post-glacial lakes such as Wular, Manasbal, Anchar, vegetatio'lal history of the Kashmir Valley especially with regard to Post-glacial climatic alterations, Dal, etc. biotic influences on vegetational development and The Pir Panjal Range - a sub-Himalayan' the phytogeography of the region. mountain - rises abruptly from the plains The sequence is divisible into eight stages. The and forms'a major barrier between the forest history began with an open vegetation with conifer woods consisting of blue pine (P.inus w(,lli Panjab and the Kashmir Valley proper. r.hiana) and cedar (C.dn,s deod(l1'a). During the It rises to an average altitude of ca. 3,750 m. ensuing period a set of warmtb and moisture de above sea level with peaks touching as high manding broad-leaved elements such as oaks, alder, as ca. 4,660 m. The region a50ve ca. birch (Betula alnoides). elm, walnut, maple and Rhus immigrated indicating a climatic trend towards 3,600 m. in particular has the characteristic increasing warmth. Later, forests comprised domi appearance of a once glaciated area with a nantlv of broacl-Ieaved elements estahlished them large number of high altitude lakes filling selves marking thereby a phase of optimum warmth the old glacially scoured bowls and cirques. which in turn was followed by the re-establishment of conifers, a fact pointing towards the return of Climate. - The present-day climate of the colder conditions. The conifer woods were mainly Kashmir Valley is considerably influenced comprised of Pinus walliehia'lla and Abies webbiana. by the position of the Pir Panjal Range. Some of the cha.nges in the vegetational sequence This is shqwn by the fact that although the are also believed to have been induced by man's direct or indirect influence on vegetation. Evidence region south of this range enjoys a sub of maize cultivation in the past has also been tropical climate, the corresponding area at brought out. The bearing of this work on the similar altitude to the north (the Kashmir history of some of the mesophytic elements such Valley) has a temperate climate. However, as Que·reus spp., Ahms and Betula alnoides, which in the summit region of the Pir Panjal Range, are today absent from the Kashmir Valley, is also discussed. the climate, leaving aside the rainfall, is more or less of the same (cold arctic' type on either side, though the snow melts much INTRODUCTION earlier on the southern slopes than on the corresponding northern side during the EOGRAPHY - The Kashmir valley summer months. The full impact of the (approximately 73" 55' to 75° 35' E monsoon in the entire Kashmir region occurs G and 33° 25' to 34° 40' N; TEXT-FIG. 1) in July when stations such as Jammu, record lies towards the north-west of peninsular rainfall as high as 305. mm. whereas Gulmarg India. It is bounded on the north and east on the lee side of the Pir Panjal Range, by the high mountains of the main Hima records barely 101 mm. and Srinagar even layan Range, whereas, towards the south less (58 mm.) for the same period (TEXT and southwest, the Pir Panjal Range divides FIG. 2). These facts reflect markedly the it from the plains of Jammu and Poonch. effectiveness of the Pir Panjal Range as a The valley is boat-shaped, with its floor barrier between the two sides of the range. lying at an altitude of about 1,560 m. and The humidity reaches its climax in the extends over an area of about 5,200 sq. km. winter months in the Kashmir Valley, a fact Its deepest part lies along the foot of the which is caused by the fall of mild winter Main Himalayas and is marked by the showers in the area. The temperature rises n 74 THE P\L\EOBOL\~IST
36°N 73 7S 76 36°N
3S ).....~. ,0 ~
..<)/..0 <'..iO ~..A
~". (C' ~ 'Yc. 'Y 33 ~. Chitto- l' C h cznab "9 ~~NGt...... SCALE :1=32rniks Gujrat • , , 74 ... ,75 76
TEXT-FIG. 1 - :Vlap showing Jammu & l as high as 38°C during summer in the plains NELLI, 1913; DE TERRA, 1938 and W ADIA, south of the Pir Panjal Range while it 1938). 0 rarely exceeds 3l 0 and 21 C at Srinagar and Palaeoecological studies in this region Gulmarg respectively. The winds in the have so far been confined to early Quater Kashmir Valley blow from the northwest, nary deposits (GODWIN AUSTEN, 1864; west or southeast and storm or cyclonic MIDDLEMISS, 1911; VVADlt\, 1919; DE TERRA, conditions arise at the height of the monsoon 1932; WODEHOUSE, 1935; SAHNI, 1936; in July .. IYENGAR and SUBRAMANAYAN(.1943); PURl, Previous War!? - The Kashmir Valley is 1943-53; NAm., 1960), while Post-glacial one of the fev" areas in the Indian subconti deposits have remained untouched. Never nent where unmistakable records of the theless, from the geophysical and historical Pleistocene' Glacial Age' have been located records in Northwest India, De Terra and and studied in detail. Four main oscilla Hutchinson (1936) inferred the first probable tions of climate have been detected (DAI- Post-glacial sequence of climate, in the alter- GlJRDIP ST:-iGH - POST-GL.-\CIAL VEGET:\TIO~:\LHISTORY OF KASHMIR VALLEY 7S so 0 +1----.-:J=------.:..F_-:.-M----.-:-----.:..A_--:-M~~J_ ___=_..J_--:-A...:.-.-----=S:._--=--OI\J D 400 f' ~- JOO // -" / " 200 100 1 . _. omm --=-=-=------__:------.c " ~-:""'''', . ------r------,'" ' ------RAINFALL ;00 ------::::-::;-:==:::~~::-=:------801 . --- _--:- . .:-::.----:.- ~_. 60 " ...... --- -' 40 ~--- 20 0 ...'------· _ 1. HUMIDITY 45° J5~ ------... ------... -"'-- ... 25° ------1S° S° -S° _15°.....1 . CO TEMPERATURE SRINAGAR (18'11-1'140) JAMMU (1'101-1'140) CULMARG (1'111-\'140) TEXT-FIG. 2 -- Normal monthlv values of rainfall, reI. humidity and temperature in Kashmir (data by courtsey of I ndia Meteorological Department). nation of wet and dry periods and also re sides these, very little information is avail corded the existence of two Post-glacial able on the large number of Post-glacial 'Ice Advances', i.e., the Fifth and Sixth. deposits in lakes, swamps and mires occuring The moraines of the latter have been shown in the Kashmir Valley. Detailed study of to lie in an intermediate position between these sites comprising their ecology, floris the terminal moraines of the Fourth Glacia tics, nature of their sediments and other tion and the recen t moraines of the existing aspects in relation to pollen-analytical glaciers. investigations have yet to be carried out. De Terra and Hutchinson (1936) and De Mukerji (1921, 1925), in a preliminary Terra and Paterson (1939) have mentioned account has provided some information of the occurrence of Post-glacial deposits in the the hydrarch succession at Dal lake (ca. Kashmir Valley in the form of river terraces, 1,560 m.). He maintained that the plant yellow loess and morainic deposits but be- communities of this lake culminate in diffe- 76 THE PAL.<\EOBOTANIST rent' climax' of vegetation in different loca as described by' Mukerji (1921,1925). The lities but that in general the plant succession aquatic plants are regularly harvested by is as follows: In the ' Aquatic Formation' the inhabitants for the purpose of building plants such as species of Potamogeton, Hydl·il floating arable plots within the lakes. la, Myriophyllum etc. dominate and consti Post-glacial peat is formed in the Kashmir tute 'Deep water Association: with Chara Valley in places which can best be termed forming a dense mat on the bed. The mires1 They are formed on the valley , Floating-leaf' or 'Shallow water Associa slopes of the Pir Panjal Range and the Main tion' is dominated by Trapa, Nelumbium, Himalayas at heights between ca. 1,600 m. Nymphaea, Eurale ferox etc. Further the and 3,500 m. The mires, which are distri , Reed-swamp Association' is dominated by buted within the forest belts, an! generally TYPha, Sparganium, Carex etc. and these small and usually measure between 10 and plants along with 'Floating-leaf Associa 30 m. in length and breadth. They are tion " form various Consocies. In the generally formed on gentle slopes, either 'Marsh Formation', Ranunculus aquatilis, singly or in successions of two or three Callitriche, Hippuris, Ceratophyllum, Juncus situated one above the other on more or less etc. constitute the' Herbaceous Marsh Asso flat terraces. On the valley slopes of the ciation ' and willows, sometimes poplars etc. Pir Panjal Range which are covered over by along with an undergrowth of perennial a thick sequence of Pleistocene Karewa clay herbs form 'Salicetum Association' and beds, the mires sometimes assume a quag , demp' land. The 'Meadow Formation' mire type of developmeht and peat is rarely is dominated by grasses and perennial herbs. developed. Both the ' demp land' and meadow stage' The mires occurring in the large number are converted into arable land by man. of glacially scoured basins occupying open The meadow stage on the eastern and patches within the forest or above the tree southern shores of .the Dal Lake is succeeded line have been occasionally found to accumu by' Gravel-slide' Formation or ' Xerophytic late fairly thick deposits of peat or organic Bushland' Formation - constituted by mud. These basins commonly known as Rubus, Rosa, Crataegus, Astragalus etc. , margs "' pathris ' and' khals ' are actually This is succeeded by , Mesophytic Bushland filled up glacial hollows or lakes generally Stage' or Shrub Stage in the Arrah valley situated on a mountain ledge or shelf en near Harwan, represented by Viburnum, closed by terminal moraines (WADIA, 1941). Prunus, Salix, Rosa, Ribes, etc. The next In that sense 'Gulmarg', 'Toshmaidan', stage is the' Pioneer Conifer Forest' which , Liddar marg', etc. form some of the best develops higher up the Arrah Valley from glacial basins or 'margs' on the northern Harwan upwards and consists of Pinus slopes of the Pir Panjal Range, while' Bod excelsa (Syn. P. Wallichiana - blue pine) Bangas " 'Sonamarg', 'Minimarg' and as the pioneer element. Few plants of , Liddarwat ' on the southern slopes of the Cedrus deodara also occur mixed with many Main Himalayas. shrubs. The succession culminates into a , Climatic climax Forest' at an elevation of POLLEN-ANALYTICAL 8,000 feet (ca. 2,400 m.) in the form of deodar INVESTIGATION forest. Swamps and mires. - Swamps are distri A. LOCATION, STRATIGRAPHY AND DEVELOPMENT OF THE MIRES buted all over the Kashmir Valley floor, covering either the shallow margins of lakes 1. TOSHMAIDAN; Mire-T: 740 31' E, such as those of Dal, Anchar, Wular, etc. 33 0 56' N (TEXT-FIGS. 1 & 3; PI. 1, FIGS. 1, 2) : or occupy the alluvial area of the ]helum The locality Toshmaidan bearing Mire-T river and its tributaries. They generally is an extensive meadow which covers an occur between altitudes of ca. 1,500 m. and 1 .'''ccording to Godwin and Com\"a~'(1939). "no 1,650m. flnd the bulk of their deposits is laid worc! at present exists in English to convev the sense as a result of flooding and deposition by the o[ the Swedish" myr ", which is a term meaning incoming river currents. Organic deposits both any kind of peat land, and at the same time are .infrequently formed at such sites. the vegetation type characteristic of such land". swamps developed a The)' proposed the use of the English word The have luxuriant " mire" in this general sense as a term embracing growth of aquatic plants which are distri all kinds of peat lands and all kinds of peat-land buted ecologically much in the same manner vegetation. GURDIP SINGH - POST-GLACIAL VEGETATIONAL HISTORY OF KASHMIR VALLEY 77 o .~oN 1~ ;0 )5' "'- 74JUf/777///7/A777J.77J777777777/TT77i-rrrrrr // / f ., /// I I \ ·vJ ~ !IS' "I?'! ~~//7~ oc::::=:::: _'.J""-.:, ',(;/)'/~3'1rt:~T}IJJ·Ift(!JI}I}4s4cI '~ ss~ " /{//{{/l//flj .. ,/ I Wi WU-&7/TllJ Yl7/ sol. f'.. r'. 0},-" {III', -\ NUP.PUP..)o>",: \ ,..- CALf ~,' " 0-- 25' ]0 INDEX CCi,q .. TOSHMAIDAN ARE A (KASHMIR) m:aMorain., ot 4th 91QciQtion fZZJ'or",lld QrlO ••• -- Footpothi I i --- Wat., shed o I 2 '3 MihfS ...~ Mir. -T (ToShtnClilolQtl) ~OGIQCjOIle,,".s TEXT-FIG. 3 - Map showing Toshl11aidan and adjoinini? wooded areas on the northern slopes of th.e Pir Panjal Ran~e. 78 THE PALAEOBOTA:"IST 7 IS 22 37 S2 67 90 1t2 I~" '\10 o IIIIIIIII I I 1 0 /,/ / / , 5~//// 5 ~ ~??:/::'/::'b.6~ .... / / //;:/ ...... : /::, /::, /::, /::, b. 6 /::,6 6 // ~ / / /"''lJ1': b. b. b. b. b. b. b. b. b. /// ?~/;j//;~?/////////////////// 10 10 1661 FOURTH GLACIAL. MORAINE ~ LAKE MUD ~ PANJAL TRAP I>::::JLACUSTRINE CLAY § GRASS PEAT ~ CONCRETIONS 100 150 200 M. TEXT-FIG. 4 - Semi-diagrammatic section of Mire-Tat Toshmaictan. area of about 10-15 sq. km. on the northern shallow mires covering the area, while it is slopes of the Pir Panjal Range in the Kashmir bounded by a small hillock towards the NW Valley at an altitude of ca. 3,120 m. above and hy a narrow channel on the SE. In sea level. It is situated under the shadow between the channel and Mire-T, lie moraines of huge massifs which form the watershed of the 4th glaciation (TEXT-FIG. 4). in the region and range between ca. 3,900.m. A level survey was undertaken along the and ca. 4,500 m. in altitude. SE-NW plane passing through the middle Immediately after the 4th glaciation in of the mire, with the bed of the channel as Kashmir, the deposition of glacial moraines the base level. Along the same line a series at various levels over Toshmaidan (DE of borings was made by means of a Hiller TERRA AND PATERSON, 1939; TEXT-FIG. 3), peat borer (TEXT-FIG. 4). After assessing blocked the free flow of water at several the depth of the peat layers at various inter places; this resulted in the formation of a vals, a point at 90 m. from the foot of the number of small lakes. In one of the many hillock, which was obviously the deepest and cirques scoured out by the fourth glaciation lay in the middle of the mire, was selected on the Toshmaidan side of the pass valleys, for final sampling. From a macroscopic a ser:ies of moraines now occupies the banks analysis of the samples, the remains of several of the present stream (TEXT-FIG. 3). The plants were recognized; for instance seeds lakes impounded behind these moraines. now of Scirpus setaceus and Carex sp., nucules of occur as extensive mires covering both Nitella and Chara, moss shoots and algal flanks of the main stream. colonies of Botryococcus and Pediastrum. Mire-T lies on the northwestern bank of A record of the stratigraphy, built up the stream and measures about 180 metres from laboratory examination of macerated from SE to NyV and more than 200 metres samples, is given below: from NE to SW (TEXT-FIG. 4). Towards the cm. 0-10 Brown fibrous peat with seeds of NE and SW, it almost merges with the other Scirpus setaceus, Carex sp., shoots CURDIP SINGH - POST-GLACIAL VEGETATIONAL HISTORY OF KASl-LMIR VALLEY 79 of Bryum atp£mm~,Autacomn£um to the igneous basaltic trap which patustre and other unidentified constitutes the rock matrix of the leaf and root fragments. Pir Panjal Range. The chief mine 10-125 Brown, humified, peat with the ral constituents in the clay are same plant remains as above. chlorite, epidote and small laths 2Pediastrum ' very rare'; 2Botryo of augite and felspar. Although coccus 'very rare' to 'common'. augite is mostly in an altered state, 125-165 Light brown, humified peat with the felspars are mostly undecom abundan t shoots of Webera sp. and posed, a fact which suggests that few' fruits and seeds of Carex sp. climatic conditions at the time of Pediastrum 'very rare'; Botryo the sedimentation of the clays coccus 'very rare' to 'common'. may have been essentially cold. 165-196 Light brown, fine detritus organic At present, the basalt at Toshmaidan is mud with fruits and seeds of Carex covered by a luxuriant meadow vegetation sp.; moss shoots less common; and in view of the relatively milder climate Pediastrum' less common' to ' very prevailing at the locality, the amount of common'; Botryococcus ' very rare' weathering taking place is insignificant. to ' common'. Nitetta nucules fre Presumably in consequence of this, blue grey quently present. clays are not found in the surface layers of 196-250 Grey, fine detritus organic mud Mire-T; however, brownish clay with iron with NiteUa and Chal'a nucules. stains is found towards the hill side. Detached moss leaves common, shoots rare; Pediastrum' frequent' Development of Mire-T - The stratigraphy to ' abundant'; Botryococcus ' rare' of the mire clearly shows that it started its to ' very common '. development from a small glacial lake. This 250-280 Greyish brown fine detritus organic fact is substantiated by 1) the position of the mud with exclusively Chara nucules 4th glacial moraine separating the mire and other unidentified leafy plant basin from the channel, 2) by the deposition fragments. Pediastrum' very rare' of unstratified bluish grey clays showing to ' rare'; Botryococcus ( frequent' mostly undecomposed felspars and small to ' very cQmmon '. concretions of broken traprock in the bottom 280-324 Dark brown fme detritus organic layers and 3) the hydrosere succession from • mud with leaf fragments. Botry a limnic phase constituted by deep water COCC~tS' very rare' to (common'. plant associations to terrestrial peat forma 324-368 Grey, clayey organic mud with tion. The second fact is significant since a rare occurrence of Chara and similar deposition is seen today in the glacial Nitella nucules, moss leaves and lakes situated above 3,900 m. in the Kashmir shoots. Pediastrum' very rare' to Valley. The subsequent slow and uninter , very common' ; Botryocoeeus rupted deposition of organic mud and then , very rare' to 'common'. of peat in Mire-T suggests that the basal 368-415 Bluish grey lacustrine clay with clays may well date from the time of the broken igneous trap-rock concre recession of the last glaciers in the region. tions frequently found towards the Thus the history of Mire-T in all probability base. Both Pediastrum and Botry goes back to the beginning of the Post oeoeeus absent except in the upper glacial period in the area. The Lake condi transitional lavers; but even there, tions seem to have continued until a level they are rare. The blue grey of 196 cm. had been reached after which the clays and the concretions as re basin appears to have shallowed enough to vealed by a preliminary petrologi allow a large scale invasion by mosses and cal examination, owe their origin afterwards by sedges. Since the upper limit of the lake deposits is at a higher level than 2. Explanation of the frequencies of algal colonies: the present morainic threshold, it seems Below 5 colonies per unit area - 'very rare' clear that some erosion has taken place and from 6 to 10 "" - 'rare' that the original threshold was at a higher from 11 to 20 "" -- . frequent' level. This seems to be substantiated by from 21 to 60 -. common' "" the sudden steepening of the gradient to from 61 to 100 "" -• verv common' above 100 .... - . abundant' wards the channel between 112 m. and 180 m. 80 tHE PALAEOBOTANIST The absence of any recognizable wood or about 1,650 m. Mire-B, which lies within fruit remains of tree elements in the mire the cultivated fields NW of the village is a sediments shows that perhaps the woods small swampy area measuring about 7 )( 15 never encroached upon the site. m. (N-S & E-W respectively). It is known Vegetation- At present the mire is ecolo to have been in existence beyond the memory gically at the ' meadow stage' of develop of the villagers. The mire is fed by subsoil ment. The' Meadow Formation' is domi water seeping in from a nearby water course nated by 'grass sedge and annual-cum coming from the hill-side. The terrace sup perennial herb' communities. Common porting the mire has been reworked by the elements among these are Scirpus setaceous, villagers for cultivation on either side but Eleocharis palustris, Phleum alpinum, Poten the mire site does not appear to have been tilla sibbaldi, Ranwnculus lobatus, R. rupes disturbed. The samples did not yield many tris, Trifolium repens, Primula rosea, Caltha macroscopic plant remains and the sediments palustris and A renaria foliolosa. The two which are less than one metre deep consist mpst important species in the subdominant of soft peaty clays with a few plant remains. moss community are Bryum alpinum and They are underlain by Karewa silt which is A ulacomnium palustre. In relatively higher hard to penetrate. and drier situations awav from the mire an 3. WALANWAR; Mire-V'!; 74° 19' E, additional assemblage o(herbs dominate the 34° 15' N (TEXT-FIG. 1; PL. 2, FIG. 4): This 'Meadow Formation '. They are Taraxa locality is situated about 3 km. east of the cum officinalis, Euphorbia wallichii, Gentiana village Braman and at about the same eleva argyrophylla, G. tenella, Impatiens sp., Sam tion. Mire-W is relatively larger and com bucus, ebulus Cnicus falconeri, Dipsacus iner paratively deeper than Mire-B. In fact mis, Tanacetum longifolitm~,Inula royleana, there are two mires at this localitv situated Adonis chrysocyathtlS, Pleorogyne carinthiaca one above the other and fed by" the same etc. Still higher up these are succeeded by water channel but the lower one, the deeper either' bush land stage' dominated by Juni of the two, was selected for study. Mire-W perus communis, J. squamata, J. recurva and is about 1·20 m. at its deepest and is com Rhododendron anthopogon communitres or posed of soft peaty clays with few macro by 'white-birch woodland stage' in which scopic plant remains. The bottom sediments Betula utilis is dominant and its close asso merge with Karewa sjlt, which is hard to ciates are Pyrus foliolosa, Salix elegans, penetrate. Syringa emodii and Lonicera parviflora. On 4. DAMAMSAR(lake); 74° 25' E, 33° 52' the slopes below Toshmaidan lies the coni N (TEXT-FIG. 3): Damamsar is a glacial ferous forest belt dominated by Abies webbi lake situated at ca. 3,940 m. above sea level ana for most of the middle zone and bv near Toshmaidan. It remains bounded bv Pinus wallichiana towards the upper and snow almost all the year round and has been lower limits. Their common associates are selected primarily for the study of bottom Picea smithiana and Taxus b.accata with samples from the lake bed. The interest in occasional pockets of Pruntts cornuta, Acer this particular lake has been inspired by 1) caesium, Crataegus oxyacantha and Aesculus the treeless nature of the area surrounding indica. The undergrowth is mainly made the lake, a fact that is likely to help in the up of shrubs such as Viburnum nervosum, study of the long distance transport of tree Skimmia laureola, Rosa macrophylla, Sam pollen in the region and 2) the close proximity bucus ebulus, Polemonium caeruleum etc. of the lake to Toshmaidan mires. The lake Still lower down nearing the Kashmir Valley lies nearly 9 km. from the tree line on the Basin, Pinus wallichiana occurs either in northern slopes of the Pir Panjal Range and pure formations or in association with broad about 4 km. from the tree line on the southern leaved tree elements, such as Juglans slopes from which it is separated by the regia, Celtis alpina, Morus serrata, M. range of hills almost 300 m. high which form alba, etc. the water shed. 2. BRAi\lAN; Mire-B; 74° 16' E, 34° 14·5' N The sediments accumulating at the lake (TEXT-FIG. 1; PI-. 2, FIG, 3): The name bottom are fine bluish-grey clays with con of the locality is taken from the village cretions of broken trap-rocks, a fact which of Braman in the Baramula district of reminds one of the occurrence of similar Kashmir. The village lies about 9 km. sediments in the deepest layers of Mire-T northwest of Baramula a.t an altitude of a.t Toshmaida.n. GURDIP SINGH - POST-GLACIAL \lEGETATlONAL HISTORY OF KASHMIR VALLEY 81 B POLLEN A.NAL YSIS 'Absolute Pollen Frequency' (number of tree pollen per unit area of the slides) is shown Material and Methods bv a separate curve (APF) in the AP diagram from Mire-T. A separate diagram has also Field Worh - The various samples were been prepared giving the distribution of the taken with the help of a Hiller peat borer various species of Querws and Betula sepa equipped with a 30 cm. chamber. The rated statistically for each stage of the vege samples were collected at 4-6 cm. intervals tational sequence. in the upper layers of the deposits but lower Vegetational Developmental Stages - From down the interval was reduced to 3-4 cm. the changes in the pollen curves of the ecolo For the levelling at Mire-T at Toshmaidan gical units of vegetation in the pollen dia a 'dumpy level' was used. grams eight vegetational stages (a-h cf. Laboratory Technique - Each sample was :Mire-T pollen diagrams) have been recog broken up in a little absolute alcoholcand nized and broad conclusions regarding the then treated with 5-10 cc. of 40% HF for a climatic changes have been ·drawn. The period of 24-72 ,hours. After decanting off above vegetational stages are still tentative the HF, the material was washed with a and do not represent pollen zones, which can few c.c. of di!. HCL (2 parts acid, 1 part be considered only, when fuller data are followed~y H 20) 3-4 washings with distilled available. In view of the more or less conti water. The residue left after final centri nuous stratigraphical development of Mire-T, fuging was shaken vigorously with 5 cc. of the vegetational history revealed by this glacial acetic acid; the acid was then decant mire has been taken as standard for further eooff and the tube inverted on a filter paper. correlation of pollen diagrams from rela After this the usual acetolysis treatment was tively younger mires. carried out as detailed by Erdtman (1952). The slides were prepared in glycerine jelly C. THE POST-GLACIAL, VEGETATIONAL and ringed with molten paraffin wax. SEQUENCE All the peat samples were taken in a more PoUeh Diagrams from Mire-T (TEXT-FIGS. or less uniform quantity and were treated 5-8) - In view of the peculiar physiographic in accordance with the above standard situation of Toshmaidan, Mire-T is not only technique with as little variation as possible. in a position to reflect the influence of the For cJayey samples, however, the percentage local vegetation but is also liable to record of KOH was reduced from 10 per cent to the wind blown pollen types from the rela 5 per cent or even 2 per cent. tively lower altitudes such as the floor of the The Construction of PoUen Diagrams - A Kashmir Valley and the southern slopes of standard sum of at le~st150 arboreal pollen the Pir Panjal Range. However, it is likely per spectrum was counted for each sample. that the long distance transport of pollen Percentages were calculated on the basis of would have affected the pollen diagrams total AP count per spectrum and for one from Mire-T more in the early Post-glacial NAP diagram, on the basis of total NAP phases, when the higher altitudes might not count. The former basis has been employed have been fore3ted, than in the advanced in the case of pollen diagrams from Braman stages, when the forests spread all over and Walanwar mires, whereas the latter giving a more or less representative pollen method has been utilized in preparing the rain for the whole region. Accordingly NAP diagram for Mire-T. The pollen grains when dealing with the bottom sediments of of the local aquatic elements have been Mire-T, more value has been placed on the excluded from the NAP sum. The bryo nature of the sediment, Absolute Pollen phytic spores have been represented as Frequency (APF) and NAP/AP ratio where percentage of the AP sum. The curves as the tree pollen flora has been taken as a for algal colonies such as Pediastrum and secondarv evidence. Further in order to Botryococcus are based on their number per visualize' the reality of the situation in this unit area of the slides. Additional columns region, a blue grey clay sample was taken have been provided to illustrate the number from the bottom of a glacial-lake namely of AP and NAP (NAP in case of Mire-T only) Damamsar, which is situated near the sum counted per spectrum and the relative re mit of the range above Toshmaidan at ca. presentation of NAP versus AP has been .3,940 m. and its pollen content was analysed indicated in each cas~.Besides this, the CfABLE 1). It is seen that in the blue-grey 82 THE PAL.J\EOBOTANIST TOSHMAIDAN MIRE- T (KASHMIR) IQ60 PERCENTACES OF FOTAl TREE ~ SHRUB POLLEN ~ ~ ~ ~ ~67o~:;,,,r-ob.'o~-,.6~ ~ ~ ~o ~¢ ~ J, k .6Ido no 6io 0 .6- n b70 blo lb .b JO STAGES _'-_':_r+_~ h I " ,_.~~ g ,00 '/0 f :: "I. "0 .---../j '00 180 e/ ---C~i,J[~:.t /00 :' I~,~ 1/01 ! " ': • +:+i: -?t I e, /'0 --'.- - i 1- ~- ~~~-'-~~-'-f"~ /&0 d 180 --1'----f-.--:.,.:~....~~; JOO . .. II I, )10 'f:: ~; f t :;.1+t' c )'0 . ... ," ,1"... r'" .. >00 _,_,_ - '. i~ ~I 1+1 • )80 It ... _I - + ~ b -'+-1--- ·. I • '00 "I '. -- [~dlCONCRfTlQNI ~ ~ ~ ~ ~ ~ ~~ ~~~~~~ ,---.---,.---. IIIII z~~.~~'k'.Jo r-;r 0)0 TEXT-FIG. 5 - ,'\rboreal pollen (AP) cliagram from Toshmaidan showing the percentages of total tree pollen, absolute pollen frequency (APF) and the Bryophytic spore curve from Mire-T. clay, which is the only sediment deposited in extn.>mely low value. This example makes this lake, the pine pollen amounted to almost it very clear that open alpine arctic vegeta 85 per cept of the total tree pollen, although tional conditions in this region can only be the nearest pine tree grows at a distance of adjudged from the nature of the sediments 9 km" and 600 m. lower on the northern deposited and from such factors as relative slopes. On the southern slopes the nearest APF and NAPjAP ratios. "Vhere pine pollen pine trees are 4 km. from the lake. The is preponderant, these con~iderationshave NAP accounts for 60 per cent of the total to be borne in mind in attempting to draw land plant pollen and the APF shows an conclusions as to the local conditions. CURDlp SiNGH - POST-GLACiAL VEGETATIONAL HistORY OF KASHMIR VALLEY 83 tion of blue grey clays showing felspars in an TABLE 1- POLLEN-ANALYSIS OF A un decomposed state and from the absence SURFACE SAMPLE FROM of the formation of organic lake mud and DAMAMSAR LAKE algal remains, both appearing only in sub POLLEN GRAINS NUMBER PERCENTAGE sequent stages). The meadow vegetation, OF POLLEN IN TERMS however, had already taken its foothold and GR.\INS OF AP largely comprised of Compositae,' Centros COUNTED % permae, and Cyperaceae. AP The miscellaneous Compositae pollen grains Pinus wallichiaua 148 84·5 show a frequency as high as 4-8·5 per cent, Abies 10 5·5 Centrospermae 29 per cent, Cyperaceae Juniperus 5 3·0 11 Betula 1 0·5 per cent, Clematis ?buchnaniana 8 per Ulmus 1 0·5 cent, Gramineae 2 per cent and Artemisia Alnus 7 4·0 1 Pf'r cent of the NAP sum at the beginning Juglrms 1 0·5 of 'Stage-a'. In the later samples the Corylus 1 0·5 curves for Centrospermae, Compositae and Viburnum 1 0·5 Total AP 175 C.buchananiana fall to low values, whereas the curves for Cyperaceae and Artemisia rise. NAP The increase in the APF at the, same time Cyperaceae 68 39·0 may well indicate the progressive encroach Centrospermae 11 6·0 Artemisia 1 0·5 ment of the region by woods. Other non Gramineae 18 10·0 tree pollen types found in 'Stage-a' are Plantago 3 1·7 Typha, Tribulus type, Clematis grata, Utri Ranunculaceae 37 21·0 cularia and Plantago, Caryophyllaceae 1 0·5 V m belliferae 1 0·5 The high values for the NAP curve in the ?Rosaceae 100 57·0 beginning of 'Stage-a', provide further Boraginaceae 12 7·0 evidence for the earlier conclusion that open Total NAP 264 vegetational conditions seem to have pre Total AP+NAP 439 Ratio AP 40% vailed in the area surrounding Toshmaidan. NAP 60% The subsequent decrease of the NAP curve can perhaps be attributed to the gradual increase in the tree vegetation as a whole. Vegetational stages: , Stage-b ': Blue pine - Oak phase: (Depth, 388-372 cm.)- This stage is characterized , Stage-a'; Blue pine - Cedar pha~e:by the sudden fall of blue pine and cedar (depth, 4-05-388 cm.)- This stage is charac pollen frequencies accompanied by a gradual terized by the extremel~'high values for but steady rise of oaks predominantly Q. Pinus wallichiana pollen. Except for cedar dilatata (TEXT-FIG, 6), The absolute pollen and oak which compete to some extent with frequencies remain low. the blue pine, the pollen curves for the other In spite of the considerable decline suffered trees remain quite insignificant. There are by blue pine (from 78 per cent to 52 per cent), occasional pollen grains of Abies, Picea, it still continues to dominate. The case of Alnus, Betula (utilis) , Ulmus and Viburntttn. Quercus is quite interesting since its pollen The oaks appear to be represented by predo curve grows from relatively low values in minantly Quercus semecarptjolia and Q. dila 'Stage-a' to high values of 19 per cent, tata in this stage (TEXT-FIG. 6). Ephedra is Simultaneously Abies, Betula (utilis) and represented to the extent of 2 per cent. Corylus also start their first continuous curves In view of the overwhelming presence of during this stage and show an increased pine pollen in ' Stage-a' together with mini representation over that in ' Stage-a', Pol mum APF and the highest NAP/AP ratio, len of other tree constituents such as Picea, it seems reasonable to infer that the blue Alnus, Ulmus, ]uglans and Viburnum conti pine and the other tree constituents did not nues to be present sporadically: Ephedra occur locally at Toshmaidan during this which was significantly represented in ' Stage stage but that they occurred some distance -a', almost disappears in 'Stage-b'. away on the relatively lower slopes. At Of the NAP constituents the curve for Toshmaidan the conditions appear to have Cyperaceae 'falls from 80 per cen t to 60 per been quite cold (as also shown by the deposi- cen t toward the top of this stage and the 84 • THE P ALAEOBOTA~ISl' QUERCUS DILATATA I Q.IN(ANA bOl'0 10 -,-'- 1 / ~.....-. oP~~- /r :~..., -"_ .. Q.ILEX j i , I Q. GLAUCA b~ Q. SEMECARPIFOLIA ·0 ... ;';.';"'-'-'~-~7'" 10 ._._;_.- i 1~, SrACES • h TI bOt'0 I g 10 ,~ I 0 ~ I ;~ ~ - I ,~, I bO - I e ;~ ;====::::- ~ .. • • a: u z ;E ~ I ~ I I d ------, • 2 0 ] ~ I ~ • I c I ~ I b ~E ~ • I I a ~ ~ I ~~ <10-1') (1~-lq) (l0-)4) (l~-)q),u (0';".0) (Ii-' b) ("b-l) (l'i-l'~)U~(~00l0'"70 611ULA 611ULA POLAR AXil EXINE TH'CKNES, UTiLII ALNO"DES TEXT-FIG. 6 - I, size statistics of pollen grains of the five Western Himalayan species of Quercus; II, comparative size statistics of the sub-fossil pollen grains of Quercus and the relatiye fre quencies of the sub-fossil pollen grains of the two 'Vestern Himalayan species of Betulu in the sequence of Mire-Tat Toshmaidan. Artemisia curve along with Compositae in this stage, (2) the return of blue pine to high crease slightly but the composites decline pollen frequencies once again, (3) the marked towards the close of the stage. Except for rise of the APF, (4) the appearance of algal a small rise towards the top of the stage the colonies such as those of Pediastntm and curve for Centrospermae shows uniformly Botryococcus which cOntribute, along with low values. The curve for Gramineae starts other plant constituents, towards the forma its first regular curve towards the close of tion of organic lake mud from the beginning this stage. Besides these, the other poorly of 'Stage-c' and (5) the disappearance of represented non-tree pollen types are Plan blue grey clays. tago, ferns, Typha, Dipsacus inermis and The blue pine pollen frequencies gradually Caryophyllaceae, the last two having only rise and except for a short temporary decline one grain 'of each while TYPha values remain in the middle of this stage they on the whole less than O'5 per cent. maintain high values. Cedar is fairly well , Stage-c ': Blue pine phase; (depth 372 represented throughout this stage. Both 280 cm.)- This stage is characterized by (1) A hies and Picea show more or less continuous the fall 'of Quercus and its continued uni curves throughout' Stage-c', though their formity of low values maintained throughout frequencies never rise to more than 6 per CUR-DIp SINGH _ POST-GLACtAL VEGETATIONAL HIS'rORY OF I TOSHMAIDAN MIRE-T (KASHMIR) Ig60 P[RUNIAC,I Of rOIAL NON- IR" POLLEN (ALCUlAlED 'N lERMI Of NAP ~""I~~ ~ ~T~~~r--o~:L~:;()~~:;o:;;o~~~roO~~~~~:;t~~I~'~:c.~.lo~IA(.{~ z~~~~.:,,~ to h ;~.- - '0 • 1 I -'.1-1>:-;(1 q be I /" I -'~_I:_.~:~~,~,_.,..--,---_ 1 ".: • II I ,::1[;1 1/1 " 11 I t I., ... ' . I/O '. lol I 0/ , :--.I~-~_I_~ ~. ... :r "O~III /~OlI~I _ 1.: I I I II == ::... 1 •I• '60 /01 I 1 = .. + .. 1 e, ~:I.I.~':t + .. ~ '00 111~11:1 .. 1 ...... I. + • t -----1-1: --(--1; -i-;------: .... t + : I .. :.. t of : I I' + t .. e, __ '_:_~I~: 1 + .. I .. 't .. )+ • t .. d -,!...... !._.t 4', I : r I·t I. ( ", -,,' ·1•I I --,~~ •• -r • t '--,1 ; b ., .'...... , . a 1'["'1'0---1 ~ ~~ ~blo~nn~~bI,6!,61!Ir.r;b7o~&7o~b1~~~ 16 ~~i~_~.Jo~ o 10 ~ ~ NA~P ~ ct <:( RATIO ~ :::> 5 ~z ~ UJ'i~~ -J a:~ Z z< ....>-<%ct c... ;r Ow z~.;;; .. o ~ TEXT-FIG. 7 - Non-arboreal pollen (NAP) diagram from Toshl1laidan ~Tire-T.showing the per centages of total non-arboreal pollen and tile N API AP curve. A key to the stratigraphic symbols is given in Text-fig. 5. cent. Ephedra is well represented through to have received a short impetus at the cost out and attains values reaching as high as of conifers during the middle of this stage 6 per cent. Among the broad-leaved trees, and simultaneously the NAP elements such the Quercus pollen 'curve, largely represented as Artemisia, and Gramineae, also seem to by Q. semecarpifolia and Q. dilatata (TEXT have had an added advantage to show in FIGS. 5, 6) continues more or less uniformly crease in their relative frequencies. On the throughout but, Alnus, Betula (utilis), Cory whole in ' Stage-c " Centrospermae increased lus and Viburnum have more or less inter to its maximum development whereas Cype mittent curves. The rest of the tree curves raceae and Compositae showed an overall remain more or less sporadic. However, decrease from t he previous stages. Lemna almost all the broad-leaved tree elements made its first appearance in this stage and such as Betula (utt"lis), Ulnws, ]uglans, Typha is increasingly represented through Corylus, Vibur1tum, Salix and Quercus appear out. During' Stage-c " as can well be seen M tHE PALAEOBOtANISi' from the gradual increase in the APF and represented while Botryococcus curve.is quite from the gradual fall in the NAP/AP ratios, high though towards the end of the stage it the tree vegetation seems to have increased declines to low values (TEXT-FIG. 8). and possibly many of the tree constituents Of the Characeae Chara dominates until had started encroaching upon the relatively towards the close of this stage where Nitella higher slopes. appears. The introduction of free floating plants , Stage-e ': Broad-leaved tree phase (depth, such as Lemna, Nymphaea, Pediastrum, 245-140 cm.)- This stage holds the most im Botryococcus and submerged plants such as portant place in the AP diagram from Mire-T Chara coupled with the formation of organic since it represents the climax of the vegeta lake mud at the site of Mire-T shows that the tional development which, having started environment at Toshmaidan had become from blue pine-cedar phase in " Stage-a " fairly suitable for the growth of these ele culminates in a mixed conIfer broad-leaved ments and for the initiation of a hydrosere forest in 'Stage-e'. in the lake. , Stage-d ': Blue pine - Oak phase (depth, Primarily, . Stage-e' is marked by the 280-245 cm.)- This stage is primarily a sudden rise of Ulmus accompanied by the transitional one and is characterized by the further decline of blue pine, as a result of fall of the blue pine and cedar curve from which, the conifers as a whole become subor relatively high levels to considerably lower dinate to broad-leaved elements in this stage. values and by the emergence of Quercus The cedar curve declines further and be which appears to be predominantly repre comes more or less sporadic towards the end sented by Q. semecarpifolia. of this stage. Quercus which occurs in this , Stage-d ' is transitional in the sense that stage is predominantly represented by Q. it represents a prelude to the next' Stage-e ' dilatata and Q. semecarpifoha (TEXT-FIG. 6). during the course of which the climax deve Apart from this the local mire vegetation lopment of the broad-leaved forest is reached shows an increased growth and aquatic plants resulting in their dominance over the conifers. such as Myriophyllum, Nymphaea. Lemna, Besides Quercus, Alnus also increases to TYPha, Chara, Nitella, Pediastrum and Botry wards the middle of 'Stage-d' but later ococcus attain their highest frequency values declines to relatively low values. The Abies in this stage. curve breaks off quite frequently but that , Stage-e ' is subdivided into' substage-el ' of Picea continues with low frequencies and e2 ' in accordance with the pattern of throughout the stage. The Betula curve the curves for the broad-leaved trees. remains at low values between 5 and 10 per ',Substage-el ': Blue pine - Oak - Elm cent while the curves for Corylus and Ulmus Birch (B. alnoides) phase: (depth, 245-200 em.) are rather more continuous than before. - At the beginning of this snbstage, the blue The Betula pollen grains are contributed by pine suffers a steep fall from 56 per cent to B. utilis and for the first time B. alnoides. 35 per cent and for the most part of this Ephedra is represented to a fair degree in substage, it has frequencies below 45 per cent. this stage. Abies, Cedrus, as well as Picea, maintain The curve for Cyperaceae falls steeply extremely low pollen frequencies. The Alnus from 56 per cent to 24 per cent and continues curve continues to maintain low values as to show uniformly low frequencies in ' Stage in 'Stage-d'. Betula, which is predomi d '. The Artemisia curve rises steadily nantly represented by B. alno·ides (TEXT throughout the stage, but has a peak in one FIG. 6) however, increases suddenly and for sample at the opening of the stage. Its fall the first time attains high values. Ulmus as well as subsequent rise almost synchronize holds an important position; its frequencies with the increase and decline of Centrosper rise from 2 per cent to 14 per cent at the mae pollen curve in ' Stage-d '. The curve beginning of the stage and the relative high for Grarnineae continues more or less uni values are maintained throughout. A conti formly throughout. Lemna pollen curve nuous curve for ]uglans begins in this sub appears for the first time at the beginning stage but its values do not increase beyond of this stage. The rest of the NAP consti 5 per cent. Corylus declines whereas Vibur tuents occur more or less sporadically. num attains its maximum development The moss spores are met with sporadically during this phase. Ephedra remains mort (TEXT-FIG. 5). Pediastrum remains poorly or less sporadic. GURDIP SiNGH - POST-GLACIAL VEGETATIONAL HISTORY OF I{ASHMiR \TALLEy 8' phase and thereafter does not rise above 10 per cent. The miscellaneous Compositae retains its continuous low curve, whereas TOSHMAIDAN MIRE-T 1960 Artemisia increases further and maintains frequencies above 40 per cent throughout 0 STAGES this substage. The curve for G,ramineae declines slightly but maintains frequencies ~I h - between 10 per cen t and 15 per cent through out. Typha pollen is more frequent, the 40- values rising to 5 per cent, which is its max q ~I , mum for the entire diagram; the curve gra - - duallv declines to low values towards the 10 1 close J of the substage. Myriophylhtm pollen appears for the first time in this substage '1\ 1'1,[100 forming a short curve with values rising a,s - high as 1 per cent of sum I I 120 the NAP and 1 1 I 0 : dwindling away towards the close of the 0 - 1 I 1 I '00 substage. Lemna remains sporadic for most I I 0/ I :1 l of the part. Pollen of herbs, such as A coni- 1- 1 II 160 __tum, Impatiens, Campanula, Polemonium and 1'1 I 1- :1 ~ e, ferns show increased frequencies. Pedias I I II ... .. I 01 I' tr'mn as well as Botryococcus, show their maximum frequency values in this substage and decline towards the close. The moss spores show a further rise in their values, e, though they still remain more or less rare. Apart from these the nucules of Chara and Nitella appear throughout the substage in I d large numbers. 'Substage-e 2 ':-Bluepine-Elm-Birch t: (B. alnoides)- Walnut phase: (depth, 200-140 C111.)- This substage is marked by ~ a sharp decline of Quercus from 23 per cent to 10 per cent and its subsequent low level. On the other hand it is characterized by the ~7/-&~r ~ c highest development of broad-leaved trees in general and for Ulmus, Betula (alnoides) I c= ~ I and ]uglans in particular. Apart from this b the substage is further marked by the rise a of Ulmus and Betula alnoides. The blue pine curve continues to fall gradually but inter o "0 .;, io R>,60 Is k><6 6000 '00 mittently and finally towards the close of cc...... ~;;- ~~ s ~~ this stage reaches its lowest level of 19 per ~~ ~ ~~ cent. Abies and Picea continue to maintain 1: ,,~ 0 ~~ ::> go U g low frequencies. The Cednts curve becomes ~ 3 0 v~ VI ~ ~cr sporadic towards the second half of this ~ 5 ~ 0: substage. e ~ ~~~ d. i ~ Quercus, after its fall, maintains frequencies ! ~ between 3 per cent and 8 per cent. Alnus does not show any marked change but TET-FIG. 8 - Diagram from Toshmaidan, show Betula (alnoides) develops vigorously along ing the distribution of Ped'iaslrul?1 and Bolrvococcus with Ulmus. They both attain high values colonies in the sequence of Mire-T. A ke5' to the stratigraphk symbols is given in Text-fig. 5. in the beginning of the substage but whereas Bet~tlacontinues to maintain them till the The curve for Cyperaceae rises very second half, thereafter attaining the highest slightly but the curve for Centrospermae values, Ulmus declines earlier. Towards suffers a sharp fall at the opening of the the top of this stage the Ulmus values again 88 'hiE PALAEOB01',·\NISt increase and the curve reaches the maximum , Stage-f ' is transitional in the same sense frequency of 36 per cent. as ' Stage-d ' but here it represents the phase The Juglans curve shows gradually ascend of reversal from a broad-leaved physiognomy ing values from a meagre representation at to one dominated once again by conifers. the beginning of the substage and attains The basis for this change is evidently laid in the maximum frequency of 14 per cent to 'Stage-I'.' wards its close. The Corylus curve does not In this stage the blue pine curve continues increase beyond 6 per cent whereas Vibur to rise along with Abies and Picea till the num declines considerably and becomes more second half, after which, the conifer group as or less sporadic, a whole suffers a temporary decline and The occurrence of A cer as well as Rhus Ulmus recovers to some extent but later it during this substage in their highest frequen again falls to low frequencies. cies for the entire diagram is quite significant, The Alnus curve becomes more or less though they are both found in rather low sporadic. Betula (utilis) continues to show frequencies. Aesculus appears towards the high values throughout this stage. CoryLls close of the substage. attains its highest frequencies for the whole The curve for Cyperaceae gradually in vegetational sequence in this stage whereas creases while Artemisia shows a correspond Viburnum though represented in a continuo:Is ing decline, although on the whole the latter curve for the first half of this stage declines still continues to hold fairly high values._ towards the close and becomes sporadic. Centrospermae, miscellaneous Compositae Aesculus remains poorly represented. and Gramineae continue more or less uni The Cyperaceae curve declines for the first formly throughout the top of the substage. half bu t recovers in the second half of the Fern spores, for the first time form a conti stage. Conversely the curves for Centros nuous curve towards the second half. Lemna permae, Artemisia, miscellaneous Compo increases to a considerable degree but dwind sitae and Gramineae show a constant rise les away towards the top along with the rest in the beginning but later they all fall off. of the aquatic elements. The Aconitum Ferns develop a continuous curve towards pollen frequencies increase but remain the close of the stage and the other NAP sporadic, elements con tinue more or less sporadically. Pediastrum shows a marked increase in the The aquatics such as Myriophyllum, Nym beginning of the substage but soon dwindles phaea, Chara and Nitella, disappear. Moss away with the appearance of the moss phase spores, however, continue sporadically till of the mire. In the subsequent stages the close of this stage. Botryococcus, though Pediastrum is met with sporadically in ex reduced comparatively, still maintains a tremely low frequencies. Botryococcus con continuous curve. tinues to he well represented but shows sharp , Stage-g ': Blue pine - Birch (B.etula fluctuations throughout the substage, Moss utilis) phase (depth, 80-25cm.)- This stage, spores attain their highest frequencies in this is marked by complete revertance to a coni phase and form a more or less continuous fer phase. This is apparently seen in the curve. Cltara is replaced by Nitella and re-establishment of blue pine forests together hereafter the latter continues till the middle with Betula (utilis) as the next important of the substage but later dwindles away, constituent. The curves for all the other , Stage-f': Blue pine - Birch (Betula uti broad-leaved tree components except Almts lis)- \Nalnut-Hazel phase (depth, 140-80 decline to extremely low frequencies. How cm.)- This stage is again a transitional one ever, Alnus increases to a considerable ex and is characterized by the reversal of the tent towards the close of the stage. blue pine curve, which now rises rapidly at Among the NAP, this stage is, character-· the cost of Ulmus which in turn suffers a ized by the increase in the Cyperaceae pollen steep fall from 36 per cen t to 3 per cent in frequencies which are accompanied by a this stage, Juglans is the second major corresponding fall in the Artemisia and Gra element which suffers a gradual fall in this mineae pollen values. Ferns and Impatiens period along with Quercus which further increase a good deal and both of them reach declines to frequencies ranging between their high~stvalues. Centrospermae, how 2 per cent and 5 per cent. Quercus here is ever, declines to its minimum values and be apparently represented by Q. dilatata (TEXT comes more or less sporadic. Typha in FIG, 6). creases sli~htlyand continues to be repre- CU~DIPSl!\'GH - POST-GLACIAL VEGETATIO:\AL H1STORY OF l\ASHMIR VALLEY B9 sented for most part of 'Stage-g' but de holds out a fair degree of promise. It conti clines tov,:ards the close of the phase. The nues to rise intermittently in the successive Ericaceae pollen frequency increases and stages, declining temporarily with the decline forms a regular curve for the first time in of high pollen-producing blue pine in ' Stages this stage and similarly Acowitum pollen curve d and eJ ' and reaches its maximum rise in is seen to increase a good deal. Both Erica 'Stage-h.' Similarly the NAP/AP ratios, after ceae and Aconitum pollen curves attain their being strongly influenced by local Artemisia highest values in 'Stage-g '. pollen in 'Stages-d, e and f,' decrease to their , Stage-h': Blue pine - Alpine fir (Abies lowest values in 'Stage-h'. The evidence webbiana)- Birch (B. utilis) phase thus obtained from the curves of both APF (Depth, 25-5 C111.)- This stage is the last and NAP/AP points out towards the same Post-glacial vegetational stage represen ted conclusion that perhaps not only did the in Mire-T and is characterized by the sudden vegetation cover extend itself progressively rise of Abies webbiana as an important unit during the various successive stages of the of the pollen assemblage. The curve for Post-glacial period but probably there was blue pine still holds the dominant position. also a gradual rise of the tree elements taking Alongside other conifers, Picea also shows place over the non-tree vegetation since the an increased development at the cost of earliest stages. broad-leaved trees in general. The curve Pollen diagmmsfrom Mires B 0- W (TEXT for Alnus, the last remnant of the broad FIGS. 9, 10). The pollen sequences from both leaved tree associations, declines to insigni Mire 'B' and ' W' must be considered 1) ficant values. Betula (utilis) , though fairly from the point of view of the altitude of their well represented in the beginning decline:; localities and 2) from the view point of the towards the upper end. Forest constituents present day vegetational zone in which these such as Betula alnoides, Quercus, Viburnuni, localities are situated. Both the mires are ]uglans, Ulmus and Corylus, either disappear situated at about 1650 m. above sea level or decline to extremely low values towards and within the Kashmir Valley vegetational the close of tnis stage. zone; the local vegetation is dominated by Whereas Viburnum, ]uglans, Ulmus and blue pine though the areas situated towards Corylus are still found growing in the Kashmir the centre of the Kashmir Valley have been Valley, the natural occurrence of Quercus, largely brought under cultivation. (Note: Alnus and Betula alnoides has not so far been the NAP pollen frequencies have been calcu reported. lated in terms of the AP sum in the The Abies pollen frequencies towards the diagrams from Mires B & W). close of 'Stage-h' though fairly high (14 Mire-B: (depth, 72-5 cm.)- In the pollen per cent), actually do not correspond to the diagram from Mire-B (TEXT-FIG. 9), the blue present day occurrence of the genus, which pine pollen frequencies dominate the whole appears to form almost 30-40 per cent of sequence reachingas high as 99 per cent but the tree vegetation of the Kashmir Valley generally range between 80-95 per cent, slopes of the Pir Panjal Range. Conversely declining to 72 per cent towards the top of blue pine is rather excessively represented the diagram. Abies which is represented in the pollen count. The reason for the dis by an interrupted pollen curve attains fre crepancy perhaps lies in the differential pollen quencies up to 18 per cent during the second productivity of the two constituents. half of the sequence but suddenly declines During this stage, the Cyperaceae curve towards the extreme top. Picea and Cedrus declines to a considerable degree and Grami pollen grains which are sparsely represented neae increases and attains its maximum for most part of the sequence increase towards value of 24 per cent. Artemisia also increases the top along with those of Abies but later but falls off towards the dose. Ferns which they soon decline away. Ephed1'a pollen were fairly well represented in the prev grains are met with sporadically. ious stage decline considerably in this Among the broad-leaved elements, Alnus, stage. Betula (utilis) , Ulmus, Corylus, Salix, Rhus, Although there is no denying the fact that Viburnum and Aesculus are very insigni the APF curve can be considerably influenced ficantly represented throughout the sequence. by factors such as the rate of sedimentation, The only broad-leaved tree that is fairly nature of sediment etc., it is interesting to common and which also increases towards see that the APF curve from Mire-T still the top is Juglans. BRAMAN MIRE- B (KASHMIR) 1960 PERCENTAGES IN TERMS Of 10TAL TREE POLLEN SIAGlS ...... ,~~~.., 1""""1"'"" ,..-. ,..... ,..., ,...., ,...,. roT r-o rw r-o ..-,...... r-1' iii iii i • r"""T'"T ~ rT'T""'" i • I I iii • ~ " ~ rt roT rw n ,-, r f1 r-o r> ~ ,.-. r---I Iii i I f i 0 ... + + ~ + + + + ".17) + + • + + In , I +~ + +r I'" • t I 1 I I I,. )} '71 r + 1'1 + +,+ + + + + + ~ -I 1 ,'".. +1 h + + + In + + 1 + +1 + I" ~[] t ..12 ~~H~~;f~. • I I + I 1..so i+ PINUS NA%p RATIO ~KAREWA _~ CLAY ~PEATY • NAP ~ CLAY ~COA"'SE o AP ~ PEATY CLAY TEXT-FIG_ 9 - Combined AI' and NAP diagram from Braman Mire-B. showing the percentages of arbor~al an:] n'm-arb:>real pollen calculated In terms of total AP and the NAPjAP curve. CURDIP SINGH - POST-GLACIAL VEGETATIONAL HISTORY OF KASHMIR VALLEY 91 The AP curve shows a progressive decline gram, on the whole show a rising trend in towards the top of the sequence but on the relation to the lower part of the diagram. whole continues to dominate. The progres Mire-W: (depth, 120-5 cm.)- The pollen sive decline of AP curve is, however, sugges diagram from Mire-W (TEXT-FIG. 10) shows tive of a successive clearing away of the a brief blue pine - oak phase at the bottom woods. which abruptly gives place to a mOJe or less Among the NAP constituents, the Cypera pure blue pine phase. The latter continues ceae curve generally has values below 24 per to dominate throughout the rest of the vege cent, but in the middle of the sequence, it tational sequence. Among the other tree mounts to ] 30 per cent. The curve for elements Abies is represented to the extent Gramineae after a low start gradually rises, of 4 per cent during the early phases but but after shooting up to ]24 per cent in the later has very low frequencies. Cedrus has middle, declines, but on the whole maintains low values before the middle of the profile, fairly high frequencies in the second half of and does not rise to any significant degree the sequence. Although the Gramineae till the top of the sequence where the curve pollen curve has a fluctuating course, its rises abruptly. Picea is well represented at general representation remains fairly high the base but later, it reduces to low fre throughout. Except for a brief appearance quencies. A lnus disappears altogether after at the base, maize pollen is not found until a brief sporadic record. The ]uglans pollen the second half of the sequence where the curve makes its regular appearance a little frequencies rise up to ]9 per cent but they before the middle of the sequence and but soon decline towards the top. The A rte for one interruption towards the top it conti misia pollen curve only appears towards the nues in low values throughout, rising only middle of the sequence and rises to 32 per towards the top to 6· 5 per cen t. Quercus cent synchronously with the maximum fre (dilatata) is represented to the extent of quencies attained by Cyperaceae and Grami 37 per cent in the beginning hut it suddenly neae. Miscellaneous Compositae are fairly falls to 3·5 per cent and vanishes beyond the well represented along with Plantago and middle of the sequence. Other tree pollen Centrospermae throughout the sequence. is sporadic. Salix, however, though sparsely From low values at the base, the fern curve represented shows a continuous curve to rises up to 25 per cent in the first half but wards the top. declines gradually to low frequencies in the Gramineae, in spite of the fluctuating curve, second half. Nymphaea, Lemna, Ranun is the most frequent pollen type. Though culaceae, Dipsacus, Polemonium, Caryophyl present from the base, maize does not have laceae, Umbelliferae, Polygonaceae, and a continuous curve until the middle of the Labiatae grains occur only sporadically. sequence, after which its frequencies rise as Impatiens pollen grains, however, are fairly high as ]5 per cent. Artemisia pollen fre frequent during the upper half of the sequence quencies mount up to some extent in the and those of Liliaceae are fairly abundant middle, but in the rest of the sequence they towards the top. Bryophytic spores are remain low. Centrospermae pollen on the commonly encountered during the lower whole, does not increase beyond 8 per cent half of the sequence. and almost disa ppears towards the close. It is interesting to see that the first inter Fern spores, though sparsely represented in ruption suffered by the Abies curve is syn the beginning, later maintain fairly signi chronized bv the increase in Gramineae, Mis ficant values. The rest of the NAP ele cellaneous Compositaeand ferns, and similarly ments remain more or less sporadic through the second fall is accompanied by a tremen out the diagram. The presence of the dous increase in Gramineae, Cyperaceae, aquatics such as Typha, Nymphaea and Artemisia, Plantago and Centrospermae. Fur Lemna is marked. ther up, the sudden decline in the blue pine The high proportion of AP is suggestive is again synchronized with the increase first of the continued dominance of the woods in Gramineae and later by maize. Collec over the non-tree elements throughout the tively the NAP frequencies gradually mount vegetational sequence. up and reach as high as 77 per cent in the Correlation oj the Pollen Diagrams - The lower middle portion of the sequence. In overall dominance of blue pine in the dia spite of a sudden fall immediately afterwards, grams from Mires-B and W is in line with the frequencies in the upper half of the dia- the features of 'Stage-h'. However, the / WALANWAR MIRE-W (KASHMIR) 1960 PtR([NTAGES IN TERMS OF TOtAL TPH rOLL tN STAGES ...0 '97 + r + 19) + + + 17) + 188 ~ ~ + T I I 77 ,: + + 169 + + + I -t of I 87 + + .00 I I I .05 "I- + 8' ~I-tl 89 I .0 + ., + .. 8 • t + I 7 + .. + + I I+ I +1 I 9 r .. h 78 + '0 ~ 77 + :.' 1" 1+ I 1 90 + + + + 90 1+ (, + I 1 I + ••7 I 8. II..... 228 214 .. I .' ." + + + ,.., n. r-t " l'"""""T"" .., r-t'h--n rl " " " rr rl n " "" 'i' iii i , I 0'00'00'00'00 20 '0 ,oQlO o'Oo'06'OO'OO'00'Oo'OOIO O'OOI0 0 .00 20 '0 .0 '0 '00 V" ....a:: ~ PINUS ....< w NA%p RATIO u ;i, w « TEXT-FIG, 10 - Combined AP and NAP diagram f~om \;Valanwar Mire-W, showing the percentages of arboreal and non-arboreal elements calculated in terms of total AP, and the NAP/AP curve. A key to the stratigraphic symbols is given in Text-fig. 9. GURDIP SINGH - POST-GLACIAL VEGETATIONAL HISTORY OF KASHMIR VALLEY 93 excessive representation of Quercus in the PINACEAE earlier phase of Mire-Wand the absence of Abies and Betula utilis phase at both sites Pinus wallichiana 'Blue pine '- This sub appears incompatible with the characteris fossil pollen type i3 overwhelmingly repre tics of ' Stage-h' at Toshmaidan. Consider sented in ' Stages-a, b, c and d ' and in the ing the altitude of both the mires and their ultimate stages of the Post-glacial in the situation, far away from the high altitude Kashmir Valley, when it reverts back from its vegetation, it seems reasonable to believe lowest frequencie3 reached during' Stage-e '. that blue pine was locally so abundant that P. wallichiana commonly known as ' blue Abies and Betula utilis, both of which grow pine' is the only natural pine repre3ented in in the higher regions, could not be represent the Kashmir Valley and grows abundant ed to any significant degree. Nevertheless ly all along the Northwestern temperate the case of Quercus (dilatata) is difficult to Himalayas from ca. 1,800 m. -to 3,600 m. understand. Its presence in Mire-W to a overlapping the whole coniferous belt except high degree in the beginning and then its the lower P. 1'Oxburghii zone and extends well abrupt fall without any synchronous effect above the Abies webbiana zone. It is seen on the NAP goes to show that perhaps that blue pine is more frequent at the upper Quercus was not widespread but occurred and lower limits of the range than in the locally in pockets which were eliminated in central part where it occurs mixed with other due course of time by the increasing en conifers and broad leaved elements. In croachment of blue pine. Alnus also seems Kashmir it does not ascend beyond ca. 3,300 to have suffered the same fate and both m. where it occupies the banks of ravines. the3e trees are no longer to be found in the The sites occupied by blue pine tend to be valley. The presence of aquatic elements those with conditions which are relatively in r,'Iires-B and W, which are absent in xerophytic from various causes and more , Stage-h ' of the Toshmaidan pollen sequence over, it is a pioneer colonist on newly avail is again probably due to the low altitude of able sites such as gravel deposits along the these mires in comparison to Mire-T. The streams, landslips, avalanche paths etc. aquatic elements represented in these mires (Champion 1936). In Kumaon dry situa normally grow at these elevations in the tions along with winter snow fall favours Kashmir Valley but at higher altitudes they its growth (Osmaston 1927). do not exist today. Picea smithiana ' Spruce '- The subfossil grains of P. smithiana are common but their frequency curve at no stage becomes pro NOTES ON PLANT REMAINS minent. The pollen type is found almost throughout the Toshmaidan vegetational The plant remains recovered from the sequence. Except for a few short gaps, its various Post-glacial deposits of the Kashmir curve continues to fluctuate within 6 per Valley are described here with special refe cent, rarely exceeding this value. In the rence to their individual distribution in the Walanwar pollen sequence, it is significantly vertical sequence of the deposits, giving in represented at the base of the profile, but important cases their present day ecological later, it follows an intermittent course. In distribution in the Indian subcontinent. Mire--B it is sporadically met with. The majority of the plant remains are P. smithiana is one of the two species repre pollen grains and spores. Others include sented in India and apart from its distri moss shoots, nucules of Chara and Nitella, bution in the Kashmir Valley between ca. Pediastrum and Botryococcus colonies etc. 1,800 m. and 3,000 m., it occurs widely Not many forms of seeds have been en throughout the N.W. Himalayas up to ca. countered except for those of Scirpus seta- 3,300 m .. In the Kashmir Valley it grows r;eus and Carex sp. - in association with Pinus wallichiana and Every attempt has been made to identify Abies webbiana never exceeding 3·5 per cent; each of the plant remains met with, by direct generally much less (SHER SINGH, 1929). comparison with its closest living represen The other species P. spinulosa grows only in tatives in the herbarium or sporothek. Eastern India. \Vherever there is uncertain ty in the exact Abies webbiana 'Alpine Fir' - Sub-fossil identific;ation, an interrogation mark pre grains of A. webbiana are common but their cedes the name ascribed. frequency curve does not sh'ow any marked 94 THE PALAEOBOTA?- rise until 'Stage-h'. In this stage, A. GNETACEAE webbiana becomes subdominant to Pimts wallichiana. Ephedra spp.- Four different subfossil Jn the Kashmir Valley Abies webbiana pollen types pertaining to Ephedra are met grows in almost pure formations along with with in the Post-glacial sediments from the Taxus baccata and Picea smithiana in the Kashmir Valley, but in view of the absence middle ranges. Towards the upper and of proper pollen morphological records of the lower limits it grows with Pinus wallichiana Indian species, specific identification was and ranges between ca. 2,100 m. and 3,150 m. not possible. on the valley slopes of Pir Panjal Range. Grains of Ephedra are frequently met with In the N.W. Himalayas it extends up to in the Toshmaidan pollen sequence in ca. 3,600 m. whereas in the E. Himalayas it , Stages a, c and d' up to 6 per cent. but ranges between ca. 2,550 m. and 3,900 m. later they become sporadic. They are found The growth of Abies webbiana is commonly only rarely in the Braman and \ValanwClr associated with extremely cold temperate profiles. climate with sufficient winter snow fall. Ephedra usually occurs in inhospitable Ab1:es webbiana apparently does not function arid conditions and can withstand long as a pioneer species, but comes into the periods of drought and extremes of tempera forest when it is approaching climatic stabi ture. lity (Dudgeon, 1925). Cedrus deodara ' Deodar' - Subfossil RANUNCULACEAE grains of Cedrus deodara are met with in significant frequencies during 'Stages a, b Clematis ?grata - The frequency of this and c' after which they decline but conti pollen type only becomes significant in nue in relatively low values until' Substage , Stage-f '; otherwise it remains extrpmely e1 '. Subsequently they become sporadic. sporadic. In the vegetational sequence from Mire-T, C. ?buchananiana - This pollen type is its high values synchronize with the rela first seen in ' Stage-a' but is more regularly tively dry and arid climatic phases. Pinus met with in 'Stages-f, g and h' of the wallich£ana and Ephedra happened to be its Toshmaidan vegetational seiJuence. chief associates during these phases. In the Aconitum ?lycoctomtm - This subfossil \Valanwar profile cedar increases towards pollen type is not usually met with till the the top of the sequence. In Mire-B, how- beginning of ' Stage-e ' after which it is quite ever, it remains sporadic. . frequently seen but follows a discontinuous Deodar which is a native of Himalayas is curve for a considerable period. In' Stage almost absent on the Kashmir Valley slopes h ' it is again rare. of the Pir Panjal Range and grows only in The Ranunculaceae pollen grains in general a few isolated patches. Its absence from occur only sporadically in the profiles from the valley slopes of the Pir Panjal Range is both Braman and \Valanwar. 'ascribed by Sher Singh (1929), to high humi dity in the region in winter which is un NYMPHAEACEAE favourable to its normal growth. Deodar commonly grows throughout the Nymphaea ?stellata - Subfossil grains of Western Himalayas between ca. 1,200 m. this species are mostly met with during and 3,300 m. but most commonly from , Stages-d and e '. Subsequently they cease ca. 1,800 m. to 2,400 m. Brandis classifies to be represented in the Toshmaidan profile deodar forest~into three groups, the first in but, continue to occur sporadically in ' Stage the dry arid zone of the inner Himalaya, the h' of the Braman and Walanwar pollen next covering the intermediate ranges and sequences. , valleys and the last on the outer ranges, 'N. stellata grows at pre~entin the Kashmir where the full strength of the monsoon is Vallev in the Dal and other lower altitude felt and 'where the rate of growth is faster lakes~Outside the Kashmir Vallev, this than ill forests nearer the main ramp of the species grows commonly througho~tthe Himalaya (Pearson and Brown, 1932). Ac warmer part~of India. In the present day cording to the above authors, excessive rain climate, it is incapable of natural growth in fall may be responsible in certain localities the Kashmir Valley at elevations at which for the total absence of this species, Mire-T if, situated (3120 m.). GURD!P S!NGH - POST-GLAC!AL VEGETATWNAL H!STORY OF KASHMIR VALLEY 9S CARYOPHYLLACEAE maximum frequencies in 'Substage-e 2 ', after which it declines. As many as five different subfossil pollen Acer caesium still grows in the Kashmir types referable to the family Caryophylla Valley in small patches between ca. 2,100 m. ceae are found in the Post-glacial sediments, and 2,700 m. and is widely distributed but at no stage of the vegetational sequence, throughout the W. Himalayas. the curve for Caryophyllaceae appears to be of any significance. Caryophyllaceae pollen ANACARDIACEAE types appear first in 'Stage-b' and there after occur sporadically in all the subsequent Rhus sp.- This subfossil pollen type first stages of the Toshmaidan pollen sequence appears in 'Stage-c' and attains its maxi excepting' Stages f and h '. In the Braman mum values in 'Stage-e', after which it profile, however, they are sporadically met declines. with throughout. Rhus is represented by several species in the-W. Himalayas mostly distributed, in the ZYGOPHYLLACEAE subtropical to subtemperate regions. Tribulus type - This pollen type is met ROSACEAE with extremely rarely in 'Stages-a and c' and is not seen thereafter. The members of this family are very The different species of Tribulus are dis scantily represented in the subfossil pollen tributed in the arid regions of the Indian records, although their modern distribution subcontinent. in the valley of Kashmir is quite significant. So far, occasional pollen grains of Potentilla ?argyrophilla, Pyrus ?pashia and P. ?foliolosa GERAN,IACEAE have been encountered sporadically in the Kashmir Valley Post-glacial sediments. Impatiens spp.- Two different types of subfossil pollen grains pertaining to this genus are commonly seen, having either HALORAGAE 3 or 4-colpate condition. In the Toshmaidan Myriophyllum ?spicatum - This subfossil diagram they are first seen in 'Stage-c' pollen type is first encountered in 'Sub after which they continue more or less spora stage-e ' in which its curve reaches its dically throughout the vegetational seq uence, 1 highest values. Later it declines but con significant freq uencies attaining apparently tinues sporadically until the beginning of in 'Stages-e! and decline in g'. They , Stage-f' after which it totally disappears. ( Stage-h ' of the Toshmaidan pollen sequence M. spicatum commonly grows on the floor are well represented in but the Braman of the Kashmir Valley in Dal and other profile and to a lesser extent in the Walanwar similar lakes at ,500 m. In the pollen sequence. about.l present day climate the species is incapable of existing in the Kashmir Valley at the alti SAPINDACEAE tudes as high as Mire-T (2,120 m.). Outside Kashmir the species descends to ca. 300 m. Aesculus indica - The subfossil grains of along the northern boundary of the Panjab A. indica, rarely occur in significant propor (HOOKER, 1879, p. 433). tion. In the Toshmaidan AP diagram, they are first seen in ' Stage-c " then in ( Stage-e ' UMBELLIFERAE to some extent, but later in ( Stages-f and g , they are more common. The subfossil'pollen types referred to this A. indica is widely distributed all over the family occur in extremely low values through W. Himalayas between ca. 1,200 m. and out the vegetational sequence. They are 3,000 m. In the Kashmir Valley proper it first encountered in (Stage-c' and occur. generally grows between ca. 1,500 m. and sporadically in ' Stages-d and f • of the Tosh '2,250 m. maidan pollen sequence. They occur spora Acer ?caesium - This subfossil pollen dically in (Stage-h' of the Braman and type first appears in ' Stage-d ' and gains its Walanwar profiles. 96 THE PALAEOBOTANIST So far three different pollen types have four types were encountered and (2) miscel been recognized, but in view of the stenopaly laneous Compositae. nous character of the pollen grains in the Arle111.isia- The subfossil pollen grains family no attempt has been made yet to refer of Arte1n1:sia playa significant role in the the fossil types to any definite genera or Post-glacial NAP diagram from Mire-T. species. The Artemisia pollen curve rises to its maximum extent and importance between 'Stages-d and and is relatively sparsely CAPRIFOLIACEAE f' represented in the early and later parts of Viburnum ?nervosum - This subfossil the Post-glacial vegetational sequence. This pollen type occurs quite frequently. It 'is seems to be explained by the fact that most first found in 'Stage-a' in low frequencies of the 13 species of Artemisia represented reaching up to 4 per cent in ' Stage-b ' after in the Kashmir Valley are now confined to which it declines but reappears in the next the drier alpine areas occupying high alti stage, where it forms a fairly continuous tude forest less slopes. It is likely that A rte curve. Afterwards it occurs more or less mis1:a got an added impetus for luxuriant during Post-glacial sporadically up to the middle of 'Stage-d ' growth the warmth but thereafter it rises rapidly and attains its period at higher altitudes. Moreover the low altitude species of Artemisia, now in maximum frequencies during' Substage-el '. In the subsequent stages the subfossil pollen habiting the forestless floor of the Kashmir type declines and only occurs sporadically. Valley also may have ascended the higher V. nervosum occurs with broad-leaved tree slopes during the warmth optimum and associations as well as with pure conifer for occupied high altitude meadows. These mations such as is the case at present in the two factors may have resulted in the in Kashmir Valley where it grows as a shrubby creased frequencies of Artemisia pollen in undergrowth of blue pine and I1bies webbiana Mire-T during this period. between ca. 2,100 m. and 3,150 m. Outside, The decline of the Artemisia curve during , Stages-g and h ' is difficult to explain. it is dIstributed all over the temperate Hima rather layas up to Sikkim between ca. 3,000 m. and It may possibly be that the pastoral activity 3,900 m. of man caused the reverses in the Artemisia curve but since Artemisia is poisonous to grazing animals this seems unlikely. It VALERIANACEAE seems possible, however, that since there is a simultaneous in'crease of fern spore frequen Valeriana sp.- This subfossil pollen type cies, the conditions may have been wetter in is extremely rare and so far only a single grain Stages-g and h and that this may have retard has been recovered, from 'Stage-c'. ed the growth of Artemisia. In the pollen diagrams from Braman and vValanwar, Artemisia is not well represented. DIPSACEAE In the Braman profile, .however, the Arte misia curve shows a significant rise along with Dipsacus inermis - This subfossil pollen Cyperaceae, Gramineae and Centrospermae type occurs sporadically in ' Stages-b, c, d, g in the middle of the sequence, at the same and h ' of the Toshmaidan pollen sequence. time as the sudden fall of Abies. is also met with in low frequencies in the It Miscellaneous Compositae - Among the Braman profile. subfossiI pollen grains ascribed to this group D. inermis grows throughout the tempe are, one pollen type of Tribe Cichoreae, two rate Himalavas from Kashmir to Bhutan of Anthimideae (apart from Artemisia) and between ca. -1,800 m. and 3,600 m. and is three of Cynaroideae. widely distributed over the Pir' Panjal The pollen frequency curve representing Range. miscellaneous Compositae begins in 'Stage a' at the high level of 48·5 per cent of the COMPOSITAE NAP, but soon declines to 8 per cent, rising once again during the middle of ' Stage-b ': The subfossil pollen grains of the family In the subsequent stages it remains extremely Compositae have been recorded under two low but continues uninterrupted for most general headings. (1) A rtemis1:a, of which of the period. It rises slightly in . Stages-c, CURDip SINGH - POST-GLACIAL V£GETATIONAL HISTORY OF KASHMIR VALLEY 97 d, f and g " and reaches as much as 10 per cally until 'Stage-f' of the Toshmaidan cent during' Stage-h' of the Toshmaidan pollen sequence. In the Braman profile they pollen sequence. It is equally well repre continue to occur extremely sporadically. sented in the Braman profile. In the Walan war sequence the miscellaneous Compositae LENTIBULARIACEAE grains are met with only sporadically. Utricularia ?minor - The subfos~ilgrains CAMP ANULACEAE of this species are extremely rare and so far only three grains have been encountered; Codonopsis ?ovata - This subfossil pollen they occur at widely spac,ed intervals in type is extremely rare and so far only two 'Stages-b and c '. grains have been encountered, both from LABIATAE , Substage-e1 '. Campanula sp.-Subfossil pollen grains of Campanula are quite frequently found after Subfossil grains referable to this family , stage-d' except for a single record from are met with sporadically in the profiles of , Stage-c, but their frequencies rarely exceed both Braman and Walanwar. 0·5 per-cent. Campanula grains occur spora dically in 'Stages-c, e, f and g '. PLANT AG IN ACEAE Plantago sp.- This subfossir pollen type ERICACEAE is at first encountered in 'Stage-a' after which it continues to occur sporadically in Cassiope The subfossil tetrads, sp.- the subsequent stages. is almost absent though slightly larger in size, resemble more It from 'Stages-f and of the Toshmaidan or less those of C. jastigiata. h' pollen sequence. is, however, well repre The tetrads are encountered first in It , Stage-e' and thereafter continue sporadi sented in the Braman and Walanwar pro files. cally till' Stage-g '. The maximum frequen cies occur during the latter stage. The subfossil grains are comparable to a with those of P. major. The Rhododendron anthopogon - This subfossil great extent pollen type is found very rarely. The only latter species grows in the Kashmir Valley between ca. 600 m. and 2400 m. two grains so far encountered are from , Stage-h ' of Mire-T. CENTROSPERMAE PRIMULACEAE Two different types of subfossil pollen grains have been ascribed to thi.s group. Primula ?rosea - This sub fossil pollen The pollen frequency curve for Centros type is exceedingly rare although P. rosea permae starts with 'Stage-a' reaching its forms a regular constituent of the alpine maximum development during 'Stages--c meadow-forming communities in the Kash and d ' but thereafter declines to low values mir Valley. The few grains met with are and almost disappears towards the close of from 'Stage-h' of the Toshmaidanpollen . Stage-h ' at Toshmaidan. It is well repre sequence. sented in the profiles of both Braman and Walanwar GENTIAN ACEAE POLYGONACEAE .Pleurogyne sp.- This subfossil pollen type is extremely rare and met with sporadically The subfossil grains ascribed to this family in 'Stages-c, e2 and f'. are encountered sporadically in the profile3 of both Braman and Walanwar. P OLEMON lAC EAE THYMELAEACEAE Polemonium caeruleum - Subfossil grains of this species are found frequently in ' Sub Daphne ?oleoides - This subfossil pollen stage-eJ '. They are first encountered in type is extremely rare and so far only one , St~e-c'after which they occur sporadi- grain has been encoun tered from ' Stage-c '. 98 THE PALAEOBOTANIST URTICACEAE moisture conditions best suited for its growth are more or less the same as \or Ulmus Sub-family: ULl\IAE wallichiana which generally grows in asso Ulmus ?wallichiana, 'Elm '- The pollen ciation with Juglans in the moist temperate frequency curve of Ulmus, after its sporadic deciduous forests. representation in the first four stages, begins to rise at the beginning of . Stage-e ' in which CUPL'LIFERAE it attains its maximum development; imme diately afterwards it declines and remains Sub-family: BETULEAE low in the subsequent stages. During' Sub stage-ez ', it appears that Ulmus gradually Betula, 'Birch' - The genus Bet1tla is re replaced Quercus and formed mixed conifer presented by two species, viz., B. utilis and broad-leaved deciduous forest along with the B. alnoides in the W. Himalayas. The third diminishing Quercus and other constituents Indian species, viz. B. cylindrostachys is con such as Betula (alnoides) , Alnus, Juglans, fined to the Eastern Himalayas only. In Viburnum, Corylus, Acer and Rhus. general no attempt has been made to distin Ulmus wallichiana still grows in the Kash guish between pollen of these species. But mir Valley on a small scale. Elsewhere, in the case of Mire-T, every fifth sample has this species grows from Indus to Nepal been re-examined with a view to making a between ca. 1,050 m. and 3,000 m. forming specific identification of Betula pollen for an important constituent of the Himalayan each pollen stage. The criteria employed moist temperate deciduous forests. It flou are (i) size; (ii) pore diameter; (iii) exine rishes either in association with Western thickness; (iv) general shape of the grain and mixed coniferous forests as in the Grahan (v) position of the aspis, whether protruded Nal and the Parbatti Valley in Panjab or is or close to the surface of the grain. By found in the western oak-fir forests of the close comparisons of the subfossil pollen with moist Garhwal Himalayas (Champion, 1936). the pollen of the three Iiving species, two The other common elm of the Western species viz., B. utilis and B. alnoides have Himalayas is U. laevigata which grows in the been recognized in the Post-glacial sediments Panjab Himalayas and in the Kashmir of Mire-T. The dominance of one species Valley at relatively lower altitudes than the over the other is depicted in Text-fig. 6, for other species. each stage. B. utilis is encountered for the first time JUGLANDACEAE in fair quantity in 'Stage-b', after which it continues to occur almost regularly in the Jttglans regia, 'Walnut '- The pollen subsequent stages. The pollen curve attains curve for]. regia holds quite an important its maximum value in ' Stages f , g and h ' place in the Post-glacial AP diagram from and declines towards the top of the sequence. Toshmaidan. Pollen of Juglans is encounter B. utilis is gradually replaced by B. alnoides ed sporadically in the first four stages, and after' Stage-d " the stage in which the latter starts occurring regularly almost simulta species makes its first appearance. B. neously with the increase in Ulmus in the alnoides attains its maximum development beginning of ' Substage-e1 '. Thereafter, its during' Stage-e ' along with the rest of the frequency curve increases gradually and broad-leaved elements. B. utilis remains rises to its maximum in ' Substage-ez '. In meagerly represented (TEXT-FIG. 6). Betula the subsequent stages, it gradually declines (utilis) grains are met with sporadically in to low values. In the profiles from Braman low frequencies in the profiles of both and Walanwar Juglans pollen though occur Braman and Walanwar. ring sporadically tends to increase in fre B. utilis is the high altitude birch which quency towards the top of the profiles. covers the whole Himalayan belt from ]. regia .still grows in the Kashmir Valley Kashmir eastwards up to Bhutan. It com and is widely cultivated for its wood, bark monly grows between ca. 3,000 m. and and fruits which are all of great commercial 4,200 m. altitude. B. alnoides is the low value. The tree is distributed all along the altitude birch of the Himalayas. It grows Himalayas up to the Naga Hills in the from Ravi eastwards, mainly in the moist Eastern Himalayas where it is found wild in sUbtropical and temperate parts at altitude evergreen climax forests (Bor, 1953). The between ca. 900 m. and 1,800 m. This GURbip SINCH - POST-GL.\CL<\L VEGE:1ATlONAL HtSTORY OF KASHMIR VALLEY 99 species is absenf in the Kashmir Valley, QUERCINEAE Hazara and the Murree Hills. Alnus, 'Alder '- The subfossil grains, Querctts - Different types of Quercus (Oak) except for their slightly larger size, come subfossil pollen grains are commonly en closest to the grains of A. nepalensis but the countered in the Post-glacial sediments, but possibility of the rare occurrence of a few while counting the pollen frequenci~sone is grains of A. nilida cannot be ruled out. liable to confuse their identity. In view of The Alnus pollen curve, after its sporadic this, all the types were counted under the representation in the earlier stages, actually generic name. But in order to get a broad starts ascending in ' Sfage-d '. It does not picture of the interplay of different types at any stage constitute a major element during the Post-glacial, specific identifica among the tree pollen. Alnus frequencies tion has been attempted in every fifth sample are fairly uniform throughout' Stage-e' in from Mire-T. Averages of the length of the Toshmaidan pollen diagram but after polar axis and exine thickness (cf. VISHNU wards again become sporadic. However, MITTRE & SINGH, 1962) were calculated for towards the end of ' Stage-g' they rise once each of the pollen stages. The resultant again to as much as 8 per cent but decline curves have been interpreted in relation to in 'Stage-h'. In the Walanwar profile the individual curves of the pollen grains of Alnus pollen grains are represented to the the five living species of Quercus in the W. extent of 8 per cent at the base but subse Himalayas (TEXT-FIG. 6). quently decline and after occurring sporadi The occurrence of Quercus during the Post cally until the middle of the sequence, finally glacial in the Kashmir Valley holds a lot of disappear completely. In the Braman profile interest, since, none of the five species of Alnus pollen occurs extremely sporadically. Que1'Cus existing in the W. Himalayas, is at Alnus represented by A. nepalensis and present represented in the Kashmir Valley. A. nitida in the Himalayas grows throughout Moreover, Puri (1945) has shown that all W. Himalayas as a primary seral type occur five species of Qttercus were growing in the ring commonly along the banks of larger Kashmir Valley during the early Pleistocene, streams and is mostly limited to sites with and he maintains that the present absence permanent water supply. Generally it in of Quercus from the area is due to altered vades newly formed shingle beds on land climatic conditions consequent on the up slips and screes and in swampy ground. It lifting of the Pir Panjal Range. has a wide altitudinal range extending from In the AP diagram from Mire-T, Quercus ca. 900 m. to 2,700 m. The genus Alnus is grains are seen in minor quantities almost no longer represented in the modern flora of from the beginning of the sequence. They the Kashmir Valley. seem to be mainly represented by Q. seme carpifoha in 'Stage-a' and Q. dilatala in Sub-family: COR.YLEAE , Stage-b ' in which the Quercus pollen curve rises to 19 per cent declining immediately Corylus columa 'Hazel '- This subfossil after\vards in ' Stage-c '; in the latter stage pollen type is first encountered in ' Stage-b ' the oak pollen was comprised of perhaps both but subsequently continues as a sporadic the above mentioned species. In' Stage-d ' constituent until' Stage-c '. After that the Q. semecarp·ljolia predominates and the curve begins to ascend and it attains its Quercus pollen curve rises to its maximum maximum rise in 'Stage-f '. Thereafter, it value. In' Stage-e' Quercus declines and declines but rises again in ' Stage-h " before keeps on declining in the subsequent stages its ultimate fall. till it almost disappears in 'Stage-h'. In C. colurt~agrows throughout the vVestern , Stages e, f and g " it appears to be repre Himalayas from Kashmir to Kumaon bet sented by Q. dilalala, though Q. se11'l.ecarpi ween 1,650 m. and 3, ISO m. as a constituent foha also seems to have been represented in of dry temperate dedicuous forests. Osmas , Stages e and g'. During the last phase ton (1922) mentions pure woods of C. colurna Q. incana and Q. ilex also seem to have accom on North aspects between ca. 2,500 m. and panied to some extent the other two domi 2,800 m. in North Garhwal and according nant forms. to him this is not paralleled in the moist In the Walanwar profile the genus Quercus forests where this species is only of sporadic is represented by predominantly Q. dilatata. C;>Ccurrence. This pollen type appears in extremely high 100 THE PALAEOBOTANIST frequencies at the base of the Walanwar AP sum towards the top' of the Braman profile but progressively declines to low fre sequence. quencies and totally disappears towards the middle of the sequence. TYPHACEAE Of the two species of Quercus found most commonly during the Post-glacial period in TYPha ?angustata _. Subfossil grains of the Kashmir Valley, Q. sernecarpifolia grows this species occur irregularly in extremely in exposed moderately moist situations at small quantities in 'Stages-a and b', but high elevations from Afganistan to Bhutan becOI;ne fairly regular in ' Stage-c' in which between ca. 2,400 m. and 3,600 m. It occurs they attain a relativel:' higher degree of pro in association with scattered Abies pindrow minence but again decline in ' Stage-d '. In or A. webbiana, sometimes blue pine and , Substage-e1 ' their fre<1uency curve deve Picea smithiana. In sheltered sites it is lops to its maximum extent of 5 per cent but replaced by Abies pindrow forest. At the dwindles away in ' Substage-e2 ' and almost top of the altitudinal range it merges into the disappears during 'Stage-f '. In the sub alpine forests of Abies webbiana, Rhododen sequent stage (Stage-g) TYPha grains conti dron and birch (B. utilis) but it often nue to occur in low frequencies but in the abuts on the alpine pastures (Champion, last stage (Stage-h) they almost disappear 1936). from the Toshmaiclan pollen sequence. Q. dilatata is the most common W. Hima However, they continue to occur sporadically layan Oak and occurs from Afganistan to throughout the vValanwar profile. Nepal, between altitudes of ca. 1,500 m. T. angustata grows throughout Northern 2,700 m. It grows gregariou~lyin all situa India from Kashmir to Manipur and South tions but seems to prefer cool and moist loca wards to Sindh and Coromandel (Hooker, lities. At its uppermost limit it merges with 1894, p. 489). At present T. angusiata does the Q. semecarpifolia forests whereas towards not grow in the Kashmir Valley at elevationc: its lower limit it occurs in association with as high as that of Mire-T (3,120 m.). Q. incana. T. ?latifolia - So far only a single subfossil tetrad has been recovered from SALICACEAE , Stagec '. T. latifolia grows in marshes all over Salix - The occurrence of Salix pollen N.W India. is more or less sporadic throughout the pollen sequence. LEMNACEAE Two species of Salix namely S. walli chiana and S. elegans are quite commonly Lemna ?minor - Subfossil grains referable met with in Kashmir. Whereas the former to Lemna are first met with in ' Stage-c ' but species grows in the Kashmir Valley zone the highest percentages are found in ' Stage between ca. 1,500 m. and 2,250 m.; the latter e', after which they again dwindle away. covers the alpine regions between 3,150 m. They.occur sporadically in ' Stage-h " in the and 3,600 m. profiles of both Braman and Walanwar. L. minor grows "throughout India" LILIACEAE and Western Tibet up to 2,850 m. (HOOKER, 1894, p. 556). In the presen.J climate, this As many as 6 subfossil pollen types are species does not appear. to grow in the referable to the family Liliaceae but in no Kashmir Valley at elevations as high as that case it has been possible to reach the species of Mire-T. . or even the genus, and as such, the subfossil grains have been recorded under the family CYPERACEAE name. The liliaceous grains are first en countereq in 'Stage-c' but thereafter they A variety of pollen types, which cannot continue to be seen sporadically up to be identified at the specific or even to the , Stage-f', after which they are not met generic level, can be ascribed to the family with in the pollen sequence from Toshmaidan. Cyperaceae. They apparently belong to a They are, however, sporadically met with number of species of Carex, Scirpus, Eleo in the. profiles of both Braman and Walan charis and other peat-forming sedge ocm war and reach as high as 12 per cent of the munities. GURDIP SiNGH - POST-GLAClf\L VEGETATIONAL HISTORY OF it<\SHMIR VALLEY tot At Toshmaidan, Cyperaceae pollen fre the factors which are most favourable to the quencies ascend to 81 per cent of the NAP growth of these plants. in 'Stage-a' but start declining gradually in ' Stages-b, c and d ' but in ' Substage-e2 ' BRYOPHYTA they again rise gradually to 75 per cent, declining immediately afterwards in ' Stage This is another group of dispersed plant f '. In the subseqllent stage, Cyperaceae remains which are not yet fully identified pollen again soars to high frequencies but but which are met with quite frequently in shows a relative decline in ' Stage-h '. Post-glacial sediments. They are represent Seeds of Carex first occur in the profile of ed either as sterile moss shoots or as spores. So far, mosses such as alp1:mm~, Mire-T in the beginning of 'Substage-e 2 ' Bryum and are present in all the subsequent stages. A ulacomn:i1~mpaltustre and Webera sp. have Seeds of Scirpus setaceous are also fou'nd to been indentified from among the many peat wards the end of ' Stage-g' and thereafter forming mosses in the Kashmir Valley. both types are present up to the top of the The relative frequency of the moss.shoots profile. has been illustrated in the stratigraphical Cyperaceae~pollen grains are found regu column provided with the pollen diagrams larly in the Braman profile but occur only and bryophytic spores· have been counted; sporadically in the Walanwar sequence. their frequencies are calculated on the basis of AP sum. GRAMINEAE Seven different spore types probably of bryophytic origin have so far been recovered. Subfossil grains referable to the family In Mire-T their first record is seen in ' Stage Gramineae have all been counted under the c " but their greatest prominence is in family name except Zea mays (maize). , Stages-d, e and f ' after which they almost Maize pollen grains were only detected in the disappear. They occur quite frequently, relatively low altitude and younger mires however, in the early phase of the Braman of Braman and Walanwar. Apart from sequence but become sporadic after the these, three other distinct pollen types of middle phase. In the Walanwar profile they Gramineae have been isolated but it has not are met with sporadically throughout the been possible at this stage to refer them to sequence. any definite genera. The value of the Gramineae pollen curve differs with the diagrams from different ALGAE areas. In the NAP diagram from Tosh Family: H YDRODICTYACEAE maidan, the Gramineae pollen curve becomes apparent only towards the end of ' Stage-b ' Pediastrum Sp. but thereafter it develops gradually and is almost uniformly represented throughout Pediastrum colonies occur abundantly in the vegetational sequence. The Gramineae , Stages-c and e1 " while in the rest of the curves in the Braman and Walanwar pollen stages they are rare; Ped·iastrum is totally diagrams have been amply discussed in absent from' Stages-a, band h '. dealing with the pollen sequence. Family: CHARACEAE FILICES The distribution of Chara and Nitella in Twelve different subfossil types of pterio the vertical sequence of Mire-Tis detailed dophytic spores have been recorded but in along with the record of the stratigraphy. view of their uncertain generic idtm tification, have all been counted the ti tie they under Family: HETEROCAPSINEAE I ferns'. Although the fern spores are encountered Botrycocws Sp. sporadically throughout the Post-glacial se quence, their frequencies rise to significant Botrycoccus colonies are absent in ' Stages levels only in 'Stages-g and h'. This rise a and b " but are met with frequently from was perhaps induced by the presumed in , Stage-c to g '; subsequently, in ' Stage-h " crease in the atmospheric humidity coupled they gradually disappear. The maximum with change over to more temperate climate, frequency is attained in 'Stages-d and e'. 102 tHE PALAEOBOTA)Jisf DISCUSSION AND CONCLUSIONS was wind transported over a long distance and that the forests from which it was de The pre;3ent investigation in the Kashmir rived occupied either the lower slopes or Valley constitutes the first attempt towards areas far removed from the Kashmir Valley. the study of the Post-glacial vegetational The local vegetation appears to have com history in India (SINGH, 1961,1962). Keeping prised Compositae, Centrospermae, Arte in view the montane physiography of the misia -and Clematis b1£chananiana. Kashmir Valley, the insufficient data avail Later, since an increase in APF and in able regarding the ecology of the forests in the APINAP ratios is found in the subsequent region and the limited number of the sites samples of this stage, it appears that blue investigated, the conclusions arrived at are pine and cedar became more common in the still tentative. area; thus presumably indicating an amelio The interpretation of the Post-glacial dia ration of the climate in the region. grams is beset with a number of difficulties. In ' Stage-b ' the increased pollen frequen Firstly the montane physiography of the cies of Quercus, Ulm1£s, Juglans, Alnus, etc. Kashmir Valley has led to the disposition of and the appearance of Caryophyllaceae and vegetation in altitudinal zones which might Dipsac1£s inermis in the pollen flora probably have shifted (upwards or downwards) with indicates that these elements began to grow the past changes in the climate. Secondly closer to the mire than before and that there in view of the fact that the Kashmir Valley was a definite improvement in the climate borders the tropics, tree vegetation must towards conditions favouring relatively more have existed not far removed from the mesophytic elements. Kashmir Valley during the early Post-glacial. Subsequently, however, the rise in the blue Consequently long distance transport of tree pine curve accompanying a corresponding pollen from the wooded lower altitudes to the decline in the oak pollen curve (Stage-c) forestless higher altitudes (such as at Tosh shows that the tree vegetation probably maidan) must have influenced the pollen changed in character resulting in the return rain considerably. However, by keeping of blue pine. Oa'k-woods appear. to have in view the factors governing the vegetational temporarily declined and pure pine-woods distribution in the valley and considering together with diminishing cedar, constituted the stratigraphy of the mires, NAP/AP the major part of the tree vegetation; this ratios, APF etc. the effect of long distance perhaps indicates a phase of brief revertence transport of pollen can be minimized to a tQ colder conditions. Nevertheless, the cli great extent and from the collective data mate in ' Stage-c' seems to have markedly the events of Post-glacial vegetational history improved in comparison to ' Stage-a', as is and climatic changes may be brought out. evidenced by the increase in APF and the Post-glacial Vegetational History and Clinw. change of sediment from clay to organic tic Alterations- From an overall review of the detritus mud. Besides, from the gradual factual data available so far, an attempt can rise in the APF values coupled with the more now be made towards the elucidation of the regular occurrence of the pollen of such ele Post-glacial vegetational history; and from ments as Bet1£la utilis, A ln1£s, Viburnum, this there can also be deduced a broad picture Abies, Picea and Juglans; and with the entry of the major climatic alterations that may of Aescul1£s, Salix and Rh1£s pollen, it appears have taken place during this period. Of the that the tree vegetation could not have been four sites investigated the longest profile far removed from the site at this stage. comes from Mire-T and it seems likely from Locally, the meadow vegetation also appears' the local topography and the stratigraphy of to have had fresh additions to its flora as the mire that the whole of the Post-glacial indicated by the first occurrence of Impa period isrepresented. As pointed out earlier, tiens, Campan1£la, Polemonium caer1£leum, the pollen spectrum exhibited by the earliest Gentianaceae; Umbelliferae and Liliaceae part of "Stage-a' in this mire appears to pollen in the sequence. On the whole, the represent an open type of vegetation. Since herbaceous vegetation appears to have been blue pine pollen preponderates and there is dominated by Centrospermae. Among the a low APF together with an overwhelming aquatics, floating elements Lemna, Nym majority of NAP, it is possible that no trees phaea, and the algae (Pediastrttm, Botryo existed at or near the site. It seems likely cocctts, Chara) also made their first appear that the tree pollen in the earliest samples ance in the sequence in this stage. GtJ"RotP SiNGH - POST-GLACIAL VEGETATIONAL HISTORY Of' KASHivIIR VALLEy' 103 The rise of the Quercus curve' in Stage-d " phytic in comparison to the rest of the vege which may well indicate the development of tational sequence, and since all the warmth pine-oak woods, together with the increased loving elements rose to their greatest promi amounts of Alnus and Ulmus and the first nence, it seems highly probable that during occurrence of the pollen of Betula alnoides, this stage the climatic conditions were at presumably indicates a change to much their warmest. Thus' Stage-e' may per warmer conditions. The fall in APF in this haps be regarded as a period of maximum stage is probably exaggerated by the decline warmth. of pine which produces much greater quan After 'Stage-e', almost all the above tities of pollen than the oaks, which 1I0W thermophilous elements declined (Stages f, became preponderant. g and h) and the fact that the conifers The relatively high pollen frequencies of returned to their original dominance sug Ephedra and Centrospermae in ' Stages a and gests that the climate reverted to colder con c' accompanied by the high values of pine ditions in this phase and a period of decreas and cedar in 'Stage-a' and of pine jn ing warmth had evidently set in. In the ( Stage-c' show that the climate governing early part of this phase (Stage-f), Ulmus had these woods may have remained largely a sudden fall and the pollen spectra suggest dry and cold. The development of oak that pine-birch (B. utilis)- hazel woods had woods in ' Stages-b and d " however, shows an enhanced development. Juglans started that the conditions were perhaps warmer declining gradually and Artemisia increased and moderately moist. On the whole the to its maximum e:-.:tent with a corresponding sequence of vegetation from ' Stage a to d ' decrease in Cyperaceae. shows that despite a temporary set back in Later, in 'Stage-g' Corylus declined and , Stage-c', the amelioration of the climate pine-birch (B. utilis) woods seem to have was progressive and this phase can justi established. Cyperaceae increased while Ar fiably be regarded as a period of increasing temisia declined and the frequency of fern warmth. spores rose to a significant extent for the Later, in ( Stage-e " the fact that bl-ue pine first time. Similarly the pollen percentages pollen frequencies fell to their minimum, and of Impatiens, Ericaceae, Aconitum and Cam that high frequencies of oak were accom panula rose to their maximum. panied by relatively higher values of Ulmus, During the last phase (Stage-h) blue Juglans, Betula alnoides, Alnus, Corylus, pine alpine fir (Abies webbiana)-birch (B. Viburnum, Rhus, Acer and Aesculus pollen, utilis) woods appear to have established it seems that apparently moist, warmth themselves. Amongst the herbs, Compo loving broad-leaved forests had established sitae and Gramineae increased while the rest themselves in the region. With the fall of declined. the oak pollen curve in 'Substage-e 2 ', Some thermophilous elements such as elements such as Ulmus, Juglans, and Betula Ulmus, ]uglans, Rhus and Acer apparently alnoides increased to their highest yalues and persisted into' Stage-f '. However, the fact appear to have constituted the bulk of the that all the moisture-loving elements includ forests. Further addition to the local flora ing aquatics declined at the beginning of this is apparent from the first occurrence of Myrio stage; and that eleillents such as Corylus, phyllum pollen amongst aquatics and Erica Artemisia and Centrospermae had a signi ceae and Aconitum amongst the herbs. ficant rise suggests that' Stage-f' may have From the high pollen curve shown by Arte been characterized by a more or less dry misia it appears that the meadow vegetation climate in relation to the immediately pre may have been largely constituted by Arte ceding and subsequent stages, In' Stages misia during' Stage-e '. The aquatic vegeta g and h ' the rise of Abies webbiana, Cypera tion seems to have shown its maximum ceae and ferns, accompanied by the decline development at Toshmaidan during this of Artemisia and Centrospermae and broad phase as evidenced by the highest pollen leaved trees in general, may well signify the frequency values attained by TYPha, Myrio setting in of a relatively much cooler and phyllum, Nymphaea and Lemna. Similarly, moister phase. Pediastrum, Botryococcus, Chara, Nitella and In the sequences from Mires-B and W mosses reached their highest extent and im occupying the relatively lower elevations the portance. Since the assemblage of forest abundance of blue pine together with Abies elements in 'Stage-e ' was the most meso- and ]uglans and ferns is also sl1ggestive at 104 'tHE !>ALAEOBOTA::-.rrSl' the cold and moist conditions as characteris dense woods for cultivation and pastures tic of ' Stage-h '. The presence of thermo (STEBBING,1921-26). Despite the large scale philous aquatic elements such as Typha, disforestation by the early Aryans the forests Nymphaea and Lemna in 'Stage-h' in the in the north of the Panjab were still dense profiles of Mires-B and W is due to the low at the time of Alexander the Great (326 altitude of the mires; these plants nonnally B.C.). According to Stebbing (lac. cit.) the grow at these elevations in the Kashmir intensive destruction of the forests in the Valley today. country was slowly brought about by the thus on the whole, the sequence of vege constant invasions of the Central Asian tational history deduced from the pollen peoples who broughftheir flocks with them. analytical evidence suggests that there was This period culminated with the Mohamma a period of increasing warmth followed by dan conq uest of a large portion of N. W. a period of maximum warmth and this was India and the Mughals may have been res in turn followed by a period of decreasing ponsible for further clearing of the forests warmth. This finding lends additional sup and also introducing some of the exotic port to the similar worldwide scheme of Post elements in our flora. glacial climatic change put forward by von Instances of the supply of timber from the Post (1946). Kashmir Valley are known as far back as the The only other sequence of Post-glacial times of Alexander the Great who got the climatic change from N.W. India, with which supply of wood from Kashmir via the R. comparisons might be made, is that compiled Jhelum (the Hydaspes of the old Greek by De Terra and Hutchinson (1936) from writers) for construction of boats (STEBBING, geophysical evidence. But since the above lac. cit.). Later, after the annexation of authors refer only to successive dry and wet the Panjab by the British in 1849, regular periods, their data have only a partial com steps were taken to exploit the forest wealth parative value. Moreover, in the absence of of the area; and the Kashmir state (then exact age determinations of the various Post under Maharaja Gulab Singh) also com glacial events studied by De Terra and Hut menced cutting and exporting timber to chinson and of the phases described in this the Panjab markets. In the year 1866, paper no correlation between the two series the' British Government planned systematic of climatic sequences can yet be made. conservancy of the forests of the Panjab A nthropogeno'Us in/bence on Post-glacial and Kashmir. vegetation - Disforestation and the intro At present an overwhelming majority of duction of agriculture by man is closely bound the human population is settled on the with the history of the people of a region, alluvial floor of the Kashmir Valley between which according to the historical records of ca. 1500 m. and 2100 m. where most of their the Kashmir Valley, is traceable to about cultivation is carried out. At higher alti 2000 B.C. when the Aryan invaders entered tudes, neither the physiography of the the Indian subcontinent and spread out in mountain slopes nor the prevailing climate successive waves. An advanced pre-Aryan allows any cultivation. Nevertheless, the civilization existed to the south-west of the thickly foreste'd slopes above 2100 m. Kashmir Valley at Mohenjo Daro and Harapa have been and are frequented for timber situated in Sindh and Punjab respectively supply. The summit region of the Pir as far back as 3000 B.C. Besides this, Panjal Range and the Main Himalayas, are palaeolithic artifacts comprising the flakes frequented by 'Gujars' (herdsmen) every of Levallois type and the flaking tradition year during the slHlJmer months. They of Neolithic age has been described from the take their flocks and herds high up on the Kashmir Valley by Hawkes (1934, p. 7) mountains from various directions, so that and De Terra & Paterson (1939). there is but a small area of the loftiest per From the meagre records of the pre-Aryan petually-snow-covered mountains that is settlemeijts in the Kashmir Valley it may left unfrequented (DREW, 1875). These be that human influence on vegetation pre areas never constitute their permanent vious to the immigration of the Aryans was abodes. negligible. The Aryans were, however, an The Toshmaidan pollen diagrams which agricultural and pastoral people and accord represent a complete Post-glacial vegeta ing to the historical records (cf. The Maha tional sequence surprisingly yield little, if bharata) they burnt away huge stretches of any, information of human interference with CURDIP SINGH - POST-GLACICAL VEGETATIONAL HISTORY OF KASHMIR VALLEY lOS \ natural vegetation. This is due to its being tive of this activity which might have existed situated at high altitude and also since the since the earlier periods. One remote indi cultivation in this region has been confined cation favouring this possibility is the over to the valley floor. This is evidenced by whelming representation of A rtemisia at the absence of maize pollen grains which are Toshmaidan since 'Stage-d '. In view of \ler~1conspicuous in the mires at lower alti the poisonous and consequently unpalatable tudes (Mires Band W). nature of Artemisia, any increase in the The elements such as Gramineae, Plantago, grazing activity in the meadows or pastures Artemisia etc. which have been noted to be where Artemisia forms a member of the vege the immediate successors after forest cut tation is likely to result in its proportionate tings, fires and catastrophic destructions of increase in the pollen rain. forests by other means in the European dia In the pollen diagram from Mire-B of grams and have WOn the title of culture which the vegetational sequence is correlated pollen (IVEI-:SEN, 1949) behave differently in with'Stage-h " the sudden decline in forest the NAP diagram from Toshmaidan. tree elements corresponds with the sudden Whereas Plantago is insignificantly repre rise of the NAP frequencies in general and sen ted t hroughou t the sequence, grass pollen of Gramineae in particular. Here it is plays a uniformly steady role after' stage-b '. the Abies which suffers most. Its sudden Artemisia, however, after having bern spar de.cline from 12 per cen t to 2 per cent is syn sely represented in ' stages a, band c ' rises chronous with the sudden rise in Cyperaceae immensely in ' Stages d, e and f'. Its rise from 6 per cent to 130 per cent, Gramineae in ' Stage-d ' and' Sub-stage-e ' does synchro from 18 per cent to 123 per cent, Artemisia nize with the sudden fall of blue pine but in from 10 per cent to 32 per cent and Plantago the absence of any accompanying increase from 2 per cent to 8 per cent of the AP sum, of Gramineae or Plantago and moreover with Similarly, the decline in the blue pine to out any recognizable rise in the NAP curve wards the top of the profile is accompanied in general, it appears that the fall of blue by an increase in Gramineae and maize, pine in the region was mainly due to an Both the above vegetatioltftl shifts appear amelioration of climate which showed itself to reflect the activity of man in the area in in the corresponding rise of broad-leaved the recent past. The increasing values of elements. The increase of Artemisia appears ]uglans towards the top of the profile are to have. been local at Toshmaidan for it suggestive of plantations of walnut by man, alternates with Cyperaceae in 'stages d, e The pollen sequence from Mire-V", pre and f'. It declines towards the close of sents evidence of a more or less continuous , Sub-stage-e 2 ' presumably due to the ,in agricultural activity in the region as is clear creasing moisture and increases during from the regular curve of maize but the , Stage-f' suggesting a relatively' drier phase'. NAP/AP curves hardly present any evidence In the latter stage its rise is again accom of clearance. The rise in the frequencies of panied by the sudden fall of Ulmu,s - an both ]uglans and Cedrus towards the top of instance which in the absence of other accom the profile is suggestive of plantations of panying' indicators of forest clearance can walnut and cedar by man. be interpreted either in terms of climatic The discovery of maize pollen in the mires reversion from warm to cold or a widespread at Braman and '~'alanwaris rather interest disease or in terms of a decline caused ing. For a long time maize (Zea mays) has through lopping of elm for fodder (HEY been considered to have had a Mexican origin BROCK, 1963). Heybrock, who made an (DARLINGTONand JANAKI AMMAL, 1945) and intensive study of the practice of lopping this has been supported by playnological of elm for fodder in Kashmir and other evidence recently (BARGHOORN,et. al. 1954). adjoining areas gives great credence to this But according to Anderson and Stonor (1949) practice as a possible cause of prehistoric maize reached the Americas from the old elm decline, Plantago is almost absent in World. Earlier in 1945 Anderson suggested 'Stage-f' and Gramineae remained much the Upper Burma as a possible centre of origin same as before. from where maize had spread to the west Furthermore, the commencement of graz coast of the New World via Java, Timor etc, ing is also not clear from the Toshmaidan At present it is not possible to say any NAP diagram, The sudden fall of the NAP thing about the antiquity of maize in the curve in 'stages-f, g and h' is perhaps sugges- Kashmir Valley and its bearing on its origin, 106 THE PALAEOBOTAi'Hst largely because of the shallow depth of the TERRA & PETERSON, 1939; PURl, 1945, 1947). Mires Band W. However, it is hoped that The Pir Panjal Range is believed to have pollen-analytical investigations of deeper acted as an effective barrier against the mires from the valley might enable us to advancing monsoon winds into the Kashmir determine the antiquity of maize in Kashmir Valley, resulting in a change of climate un and to solve the controversial problem of favourable to Oak, Alder and Betula al its origin. noides eventually bringing about their ex Phytogeographical Considerations - From tinction from the valley. In the light of the account of the Post-glacial vegetational their high representation in the pollen pro history of the Kashmir Valley, it becomes files, the problem that confronts us is apparent that the modern distribution of the whether these elements really immigrated natural flora of this region has been largely into the valley or not during the Postglacial influenced by the changes in the Postglacial climatic optimum. climate leaving aside the artifically forested It need not be pointed out that the vege or the forestless regions created by man for tation on the southern slopes of the Pir which very little and insufficient information Panjal Range (of which Quercus and Alnus has heen obtained from the present investi form a part), must have reacted to the Post gation. The sites examined come only from glacial climatic alternations in much the the northern slopes of the Pir Panjal Range same way as the vegetation on the corre and obviously they reflect largely the vege sponding northern slopes. It is likely that tational changes along the Kashmir Valley there was an upward shifting of the bro3d slopes of this range in particular and of the leaves forest belt during the 'Post-glacial Kashmir Valley in general. However, since optimum'; it is also possible that a fraction the evidence of the vegetational history is of the subJossil pollen preserved in the based upon pollen grains, an allowance has Kashmir Valley mires might have come from to be made for the wind blown pollen coming the southern slopes of the Pir Panjal Range. either from the southern slopes of the Pir The pollen spectrum from one of the surface Panjal or from other places far removed from samples from Toshmaiclan showed the pre the Kashmir Valley. sence of both Quercus and Alnus pollen grains While the majority of the plant species to the extent of 0·5 per cent and 2 per cent represented in the profiles find their living respectively. A more or less similar situation representatives in the modern flora of the with 4 per cent Alnus and no Quercus was Kashmir Valley, there are certain elements seen in the Damamsar sample from the such as Quercus, Alnus and Betula alnoidcs summit of the range. No subfossil pollen which though represented by pollen in the of ~itherof the above two genera was recover profiles to the extent of 37 per cent, 8 per cent ed from the surface samples of Mires-B. and and 15 per cent of the total tree pollen respec W. This evidently shows that a certain tively at their maximum do not grow in the amount of wind transported pollen crosses Kashmir Valley today. Out of these, how over the range from the southem flank to ever, the first two do grow on the Southern wards the Kashmir Valley. slopes of the Pir Panjal Range but not in However, there is no denying the fact that Poonch region adjoining Toshmaidan. Five during the early stages of the Post-glacial species of Quercus occur on the southern period when an open vegetation existed on slopes of the Pir Panjal Range in Udham the Kashmir Valley slopes of the Pir Panjal pore, Rampur, Mirpur, Muzaffarabad and Range, wind transported pollen would have ]helum Valley and one species .of Alnus in influenced the pollen spectra considerably Muzaffarabad, Ramban and Udhampore (cf. 'Stage-a', Mire-T). Bu t in the later (LAMBERT, 1933). Betula alnoides, however, phases, considering the relatively high values grows from Ravi eastwards in the Panjab of these elements, it seems doubtful whether Himalayas. the presence of this pollen can be ascribed to The chief interest in the above three long distance transport. The Pir Panjal constituents centres round their occurrence Range is crossed, however, by a number of in the Kashmir Valley during the early Passes among which the lowest is the Bara Pleistocene and the belief that they were mula Pass (ca. 1560 m.) through which flows exterminated from this region due to the the main R. ]helum. The other Passes are uplift of the Pir Panjal Range towards the situated at ca. 2,800 m. (Banihal), ca. 3,420 end of First Interglacial (SAHNl, 1936; DE m. (Pir Panjal) and above. How far these GORDIP SINGH - POST-GLACIAL VEGETATIONAL HISTORY OF KASHi\{{R VALLt\' 101 passes acted as routes for the immigration suggestions so kindly given during the pre into the valley during the climatic optimum paration of this paper. for Quercus, Betula alnoides and Alnus is yet To Professor R. N. Chopra of the Botany to be ascertained. But that their role must Department, Punjab University, Chandigarh, have been considerable cannot be denied. the author is sincerely thankful for rendering It is hoped that the future work in this help in the identification of sub-fossil mosses. region will be able to solve this vexing To Dr. A. D. Kharkwal of the Geology Depart problem. ment, Lucknow University, the author is thankful for his kind help inthe petrological examination of the rock and clay samples ACKNOWLEDGEMENTS from Toshmaidan and Damamsar (Kashmir). The author is also thankful to the autho The author wishes to express his deepest rities of the Birbal Sahni Institute of Palaeo gratitude to Dr. R. N. Lakhanpal for his keen botany, Lucknow, for the award of an Assist interest and guidance throughout the pro antshrp at the Institute and to the Council gress of this work. The author is especiallv of Scientific and Industrial Research, New indebted to Professor G. Erdtman, Director Delhi, for the award of Senior Research of the Palynological Laboratory, Stockholm Fellowship during the combined tenure of Solna, Sweden, for kindly imparting the which the present work has been carried necessary practical training in the field-and out. laboratory techniques during his stay at the The author also wishes to express his sin Birbal Sahni Institute of Palaeobotanv, cerest thanks to the authorities of the Forest Lucknow, as a Visiting Scientist. The Department of Jammu and Kashmir State author is also deeply obliged to Dr. Alan for their kind co-operation during the field G. Smith of the Botany Department, Queen's, work in the Kashmir Valley and to the India University, Belfast and to Dr. Vishnu-Mittre Meteorological Department for kindly sup for their constructive criticism and valuable plying the meteorological data. REFERENCES ANDERSON, E. & STONOR, C. R. (1949). Maiez Western Himalaya. J. Indian bot. Soc. 4(7 & 8): among the Hill People of Assam. Ann. kIo bot· 233-285. Cdn. 36: 355. ERDTM.• N, G. (1952). Pollen morphology and Plant B.\RGHOORN, S. E., CLISBY, IC H. & '"VOLFE, .M K. taxonomy. .-\ngiosperms. Waltham. Alass. (1954). Fossil Maize from the valley of Mexico GODWIN, H. 8; CONWAY, V. M. (1939). The Ecology lJot. Mus. Leafl. Harv. 16(9): 229-39. of a raised bog near Tregaron, Cardiganshire. BLATTER, E. (1927-28). Beautiful flowers of j. Ecol. 27(2): 313-363. Kashmir. I & II. London. GODWIN-AUSTEN, H. H. (1864). Geological notes BOR, N. L. (1953). Manual of Indian forest. botany. on part of the north-western Himalaya.s. J. Londoll. geol. Soc. London. 20: 383-88. C)[.~~IPION,H. C. (1936). A preliminary survey of HAWKES, CHR., J & DE TERRA, H. (1934). Palaeo the forest types of India ancl Burma. Indiml lithic human industries in the north-west Punjab For. l?ec. 1: 1-236. and Kashmir and their geological significance. D.\INELLl, G. (1923-35). De Filippi Expedition to Mem. Conn. Acad. Arts Sci. 8: 1-14. the Himalaya (1913-14), Memoirs and other HEYBROCK, H. M. (1963). Diseases and Lopping publications, Vol. I-XXX. Bologna. for Fodder as Possible Causes for a Prehistoric D.\RLlNGTON, C. D. & AMMAL, E. K. J. (1945). Decline of Ulmus. A eta. Bot. Nea-rlandica 12 : Chromosome Atlas of cultivateci Plants. London. 1-11 DE TERR.\, H. & HUTCHINSON, G. E. (1936). Data HOOKER, J. D. (1872-1897). Flora of British India. on POot-glacial climatic changes in Northwest I-VI T. Londoll. India. Curro Sci. 5: 5-10. IVERSEN, J. (1949). The influence of prehistoric DE TERR.~,H. & PATERSON, T. T. (1939). Studies man on vegetation. Damna·rk$ geol. Vnders. on the Ice Age in India and Associated human Ser. 4. 3(6): 1-25. cultures. Washington. IYENGAR, M. O. P. & SI:BRAHMAN.,\YAN, R. (1943). DREW, F. (1875). The Jummoo and Kashmir Fossil diatoms from the Karewa beds of Kashmir. territories. London. Proc. nat. Acad. 5ci. India. 13(4). DUDGEON, W. (1925). The ecology of Tehri LAMBERT, W. J. (1933). List of trees and shrubs Grahwal: A contribution to the ecologv of the for the I and Kashmir State. For. Bull. Dehradun. 80: Idem (1948). The flora of the Karewa series of 1-40. Kashmir and its phytogeographical affinities MIDDLEMISS, C. S. '(1911). Sections in the Pir with chapters on the method5 used in identi Panjal Range and Sind Valley, Kashmir. Rec. fication. Indian For. 74: 105-22. geol. Surv. India. 41: 115-144. S,\HNI. B. (1936). The Karewas of Kashrr.ir. MUKERJI, S. Ie (1921). The Dal Lake (Kashmir): Curr. Sci. 5: 10-16. a study in biotic sllccession. Proc. Eighth Indian SHER SINGH (1929). The effect of climate on the Sci. Cong.; Proc. A siat. Soc. Bengal. (N.S.) 17: conifers of Kashmir. Indian For. 55: 189-203. 185-187. SINGH, G. (1961). Further Studies in the Plyno Idem (1925). The ecology of the Dal Lake Region, logy of Kashmir. Ph. D. Thesis, University Kashmir. Proc. Twelfth Indian Sci. Congr.: of Lucknow. Lucknow. 188-189 Idem (1962). Post-glacial Vegetational History of NAIR, P. K. K. (1960). Palynological investigation the Kashmir Valley. Internat. Conference on of the Quaternary (Karewa) of Kashmir. J. sci. Palynology, Tucson (Ariz.), Abstr. Pollen et Spores induslr. Res. 19B(6): 145-54. 4(2): 376-377: OSMASTON, A. -E. (1922). Notes on the forest com STEBBING, E. P. (1921-26). The Forests of India. munites of the Garhwal Himalava. J. Ecol. 10: Vol. I-III London. 129-67. VISHNU-MITTRE & SINGH, G. (1963). On the IdE'm (1927). A Forest Flora for Kumaon. Pollen of the \Vestern Himalayan Oaks. J. Allahabad. Indian bot. Soc. 42 (1): 130-134. PEARSON. R. S. & BROW:-I, H. P. (1932). Commer VON POST. L. (1946). The prospect of pollen cial timbers of Tndia. 1. Calcutta. analysis in the study of earth's climatic history. PURl, G. S. (1945). The genus Quercus in the NewPhvlol. 43: 193-217. Karewa deposits of Kashmir with remarks on WADIA. D~N. 1938. Deposits of the Ice _\ge in the oak inrests of the Kashmir Valley during Kashmir. Qual·t.]. geol. min. & met. Soc. India the Pleistocene. Proc. l ndian A cad. Sci. 22: 10(4). 191-99. 232-56. Idem (1941). Pleistocene Ice Age deposits of Idem (1945). Some macroscopic plant remains Kashmir. Proc. nat. Inst. Sci. India 7: 49-59. referred to the Betulaceae from the Karewa Idem (1953). Geology of India. London. Deposits of Kashmir. Ibid. 22: 257-73. WODEHOUSE, R. P. 1935. The Pleistocene pollen Idem (1947). Fossil plants and the Himalayan of Kashmir with an introductory note by uplift. J. hut. bot. Soc. lyengar Commemoration H. D~ Terra. lv/em. Conn. Acad. Arts Sci. 9: Fol.: 167-84. 1946. 1-18. EXPLANATION OF PLATES PLATE 1 and blue pine-alpine fir mixed woods in the back 1. Toshmaidan Panorama. ground. 2. Site of Mire-T at Toshmaidan. 4. Site of Mire-B at vValanwar with blue pine PLATE 2 alpine-fir - broad-leaved mixed woods in the 3. Site of .Mire-B at Braman with cultivated fields background. o ,-<