Landscape Ecology vol. 9 no. 3 pp 175-190 (1994) SPB Academic Publishing bv, The Hague Coincidences between different landscape ecological zones and growth forms of Cembran pine (Pinus cembra L.) in subalpine habitats of the Central Alps Anselm Kratochwil’ and Angelika Schwabe2 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Universitat Osnabruck, FB Biologie/Chemie, Fachgebiet Okologie, Barbarustr. I I , 0-49069 Osnabruck, Germany; 21nstitutfur Botanik der Technischen Hochschule/Geobotanik, Schnittspahnstr. 4, 0-64287 Durmstadt, Germany Keywords: Cembran pine, Nutcracker, multiple trunk trees, intraspecific coexistence, area of glacier recession, timberline Abstract The ‘forgotten’ depots of the European Nutcracker (Nucifraga c. caryocatactes) often ‘lead’ to the development of tufts of Pinus cembra. In many cases the other individuals of such tufts are not suppressed by the fittest one, rather there is an intraspecific coexistence up to the senescent stage of the trees. There are fusions of separate trunks, and so frequently the individual history of older trees can only be reconstructed by studying sutures, crown structures or trunk cross sections. Different types of trunk fusions are worked out. By means of transect counting the occurrence of these ‘multiple trunk trees’ is documented quantitatively in different landscape ecological zones of the Engadin region (the Grisons, Switzerland). The data base is 3024 counted microsites of Pinus cembra individuals arising from seeds, including 5272 living individuals. These ‘multiple trunk trees’ significantly play an important role in the landscape ecological zones of recent glacier recession and at the alpine timberline. Their growth forms have a higher biomechanical stability. 1. Introduction Taking as an example the growth forms of Pinus cembra we would like to point at the ‘intraspecific co-existence’ where positive interactions between individuals enhance the growth in extreme habitats. Mostly this coexistence depends on the intimate relationship between Cembran pine (Pinus cembra) and the Nutcracker (Nucifraga c. caryocatactes). To our knowledge the growth forms of Cembran pine have not yet been treated synoptically as to also include their planting by the Nutcracker, and the degree of the fusion of the trunks. Also the work about Pinus cembra in the east Alps (Schiechtl and Stern 1975- 1984) lacks this aspect. Schroeter (1908) mentions 7 different physiognomic Cembran pine types, and also Klein (1908) deals with the ‘physiognomy’ of the Cembran pine, particularly charac- terizing their bizarre shapes when growing close to the timberline. Rikli (1909) presents a compilation of different growth forms. Klein (1.c.) already recognized certain regularities in the growth form without, however, perceiving the connection to the “planting” by the Nutcracker. On the actual timberline the growth forms of young Cembran pine have been analyzed and classified as ‘jay plantations’ by Holtmeier (e.g. 1985), among other things taking into account damage through fungus attacks, long-time snowcover, or damage by blowing ice particles. We asked the following questions: - Is there a connection between the frequent occurrences of multiple trunk Cembran pine individuals and the ornithochorous dispersal? - Are there different types of trunk fusions? - Is it possible to demonstrate fusions of several 176 I I I ...... Boundary Swiss Na tion a 1 Park / I I I I \ - Rivers and Brooks \ I f ..._ Morteratsch Piz Bernina h . i. ___ '\ Frontier zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHG zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Switzerland I Italy \ -'Glacier 4049 m ,/-', /-t, )---' /0 I Areas of A\\\v zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Investigation '.. \ \ J Po&hiavo( \ - 0 II\ \ 5 lOkm Fig. I . Map of the area investigated in the Engadin, Poschiavo and in the high valley of Livigno (small map: further areas in The Grisons and Valais, location of the map section). Transects: A. Regions in front of the glacier; B. Actual timberline; C. Parklike tree stands; D. Juvenile growth; E. Mature woodland, non-managed; F. Mature woodland, managed extensively. - trunks when analyzing the trunk cross sections? Are there coincidences between certain growth forms and landscape ecological zones, and if this can be recognized: what is the reason? 2. Topographic situation of areas and methods Our investigations were mainly carried out in the lower and upper Engadin and the Poschiavo. Main 180 DartlarouD % 50 Morteratsch area in front of glacier (icefree 1887 - 1920) 50 40 actual timberline ("outoosts" -100cm) 40 I 30 20 10 30 1 2 DartlQrOUD % 10 4,3 0 7 50 40 40 n = 797 30 30 20 20 I 9,5 10 0 0 3 4 5 6 7 6 number Pinus cembra indiv./microsite 9 d 60 10 5 )art/group % C 50 2 2 3 4 number Pinus cembra indiv./microsite 1 parklike zone beneath the actual timberline 1 354 20 5 6 4 number Pinua cembra indiv./microaite 60 n n = 163 3 0 "1( 43.5 n juvenile growth up to 3 m 47f i 2 10 DartlarouD K within mature woodland 36.1 3 4 5 6 7 number Pinus cembra indiv.fmicrosite 8 9 partlgroup % 51,5 n mature woodland with p.min.partc thin stems, non-managed 4 80 70 I mature woodland with p.min.parte 64.7 thin stems, managed extensively 60 I n = 163 32.5 50 n 40 I 909 29,8 30 12.5 20 10 0 1 2 3 number Pinus cembra indivhnicrosite 4 1 2 3 4 number Pinus cembra indiv./microsite 5 zyxwvutsrqponm Fig. 7u-f. Distribution pattern of 1- and multiple trunk Cembran pine individuals in different landscape ecolo@calzones in the Engadin and Poschiavo. Data were collected when counting transects in the years 1989-1991; n designates the number of tufts counted. dividuals, giving the impression of one tree (Fig. 6). The sutures between the individuals are quite often clearly discernable. More than 10 trunks can be involved in the fusion, however, as a rule, there are less. As we shall show, the multiple trunk individuals of Pinus cembra occur quite frequently on sparsely overgrown ecological border habitats (possibility of the root extending to the side), e.g. on moraines or on rocks; they are also not absent in extensively 181 Pinu8 cembra IndivJmlcro8ite types, characterizing definable landscape ecologizyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA cal zones (see Fig. 1). * A . Regions infront of the Morteratsch glacier, 1900 m NN, trees and small treesfrom (1.5) 2 m height, 2.17 Fig. 4, 7a, 8 In the area in front of the glacier, which has been icefree since 1857-1920, Pinus cembra appears zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA 1,s +I ,4C like on the neighbouring Roseg glacier - as a pioneer and reaches at present a maximal height of 5 m. Whereas in the transect, which has been icefree since 70 years only, we discovered only two small Fig. 8. Germinated/ growing Pinus cembra individuals per trees. “nest”: different landscape ecological zones and age groups. Ludi (1958) already mentions individuals of Mean values and standard deviations are given. Cembran pine in the front of the Morteratsch glacier, which, at that time, had a height of less than 1 m. The portion of the 3-trunk individuals managed forests with a higher portion of fine with 31,3070 has the highest value of all types inground. Whether here also it comes to intraspecific root intergrowths, is not yet known. However, vestigated. A low average age and less damage by ibexes combined with a favourable light climate are judging by the results of other pine species, this important ecological factors in this area. could be expected (s. chapt. 4). A more efficacious ‘selective thinning’ will, as a rule, only enhance the growth of one individual per tuft. For this reason B. ‘Outposts’ (Holtmeier 1985) in the area of the actual timberline, 2200-2350 m NN, trees and the multiple trunk type is underrepresented in such small trees from 1-2 m height; Fig. 2, 7b, 8 forests. Regions: Above Alp Languard and Suot Paradis Since the occurrence of multiple trunk Cembran near Pontresina; Munt da la BCs-cha near Celerina; pine individuals in the different landscape ecologiVal da Camp/Poschiavo; Tamangur near S-charl; cal zones is not uniform, we would like to show Tavru near S-charl. their distribution pattern in the landscape of the In this landscape ecological zone the 2-trunk inupper and lower Engadin as an example - before we start on a characterization of the different dividuals of Pinus cembra predominate with 43,5%, however, also the 3-trunk type is well represented growth forms. with 16,7% (often occurring on rock sites); the mean number of individuals per microsite is 2,2. The area in front of the glacier and the zone of the 3.3 Distribution pattern of multiple trunk Cembran actual timberline differ by number of Pinus cempine individuals in the landscape of the upper bra individuals per microsite significantly from the and lower Engadin following types (Fig. 8). In the zone of ‘outposts’ the tufts of Pinus cemIn 29 areas (listed below under A-F according to bra display in their growth form a similarity to the the designations of the ‘Landeskarte der Schweiz polycorms of e.g. Picea abies as they have been de1:25.000’), trunk countings of altogether 5227 livscribed in the important publications of Kuoch and ing individuals of Cembran pine and 3024 microAmiet (1970) and Tranquillini (1979). Also in sites (hiding places of the Nutcracker) of Cembran North America there are corresponding clone formpine were carried out in the upper and lower Ening species near the timberline (e.g. Picea engelgadin and the Poschiavo. mannii, Abies lasiocarpa, see Holtmeier 1989). The results are presented in Fig. 7a-f and 8 and Already Campell (1950) points at the different oriare explained as follows. The areas belong to 6 182 gin of Pinus cembra groups due to the seed hiding places of the Nutcracker. C . Parklike zone beneath the actual timberline (at times overlapping and without clear demarcation) 2000-2250 m NN, Fig. 7c, 8 Areas: near Alp Languard, Muottas Muragl and Schafberg near Pontresina (without considering Cembran pine-afforestations at Schafberg); beneath Muottas da Schlarigna; Spuondas da Staz ob St. Moritz-Bad; Munt da la BCs-cha near Celerina; Alp Clavadatsch near Cristolais (Samedan); Val da Camp/Poschiavo; Tamangur near S-charl. Of this type 1-trunk and 2-trunk type occur frequently; the formerly strong grazing impact in this zone and the partially already limited light climate are surely playing a part here. In addition there may be seeds of non-ornithochorous origin. The mean number of Pinus cembra individuals per microsite is around 1,8. D. Juvenile growth up to 3 m within closed mature woodland, 1800-1980 m NN; Fig. 7d, 8 Areas: Staz Forest near St. Moritz-Bad with the exception of areas rich in Picea abies, on the fringe of Alp da Staz and Plaun da Staz, on the fringe of Val S-chura near Pontresina; Tamangur near S-charl. Countings of 1-3m high juvenile trees in the Staz Forest and boundaries and in the non-managed Pinus cembra forest ‘Tamangur’ reached mean values of 1,8 individuals per microsite. The values given by Mattes (1978), who investigated 1-20 year old juvenile trees in the Staz forest (also including dead individuals and areas strongly overshadowed by Picea abies trees: the number of counted microsites being 218) are slightly lower (1,5). Here also one has to reckon with juvenile growth originating from non-ornithochorous dispersal. It is astonishing that a comparison with nonmanaged mature woodland shows that the basic distribution of the 1-trunk, 2-trunk, 3-trunk types does not change very much. The steep slope area beneath Munt la Schera mapped by Campell and Trepp (1968) as Rhododendro-Vaccinietum cembretosum contains, next to thick stemmed Pinus cembra individuals also many thin and even dead stems. Though being part of the National Park, the envisioned virgin forest conditions are disturbed by too much deer (every young Picea abies has been bitten off). In spite of worse light conditions (in comparison with Type A-C) and the more severe competition with other tree species still 32,5% microsites of Pinus cembra with 2 trunks could be found; the mean number of individuals per microsite is 1,7. The non-ornithochorous dispersal certainly plays a part in closed forest stands. Park zone, non-managed stands and the juvenile growth type show a uniform behaviour concerning the mean number of Pinus cembra individuals/microsite (Fig. 8) and differ considerably from other types. F. Mature woodland with p.min.parte thin stems, managed extensively, 1800-2000 NN; Fig. 7f, 8 Areas: Staz forest near Pontresina, Celerina and St. Moritz-Bad; Sur Plaun near Samedan; God Spuondas-Rosatsch near St. Moritz-Bad; Tais Giuven and Chalchagn between Surovas and Morteratsch; Val da Camp/Poschiavo. Also in these closed mature woodland approximately 30% of all trees are 2-trunk Pinus cembra individuals which have already coexisted over decades. Here too, according to Mattes (1978), the non-ornithochorous dispersal is involved. The mean number of individuals per microsite is 1,4 and differs quite a bit from the other types. The distribution of trunk types according to Fig. 7f also applies to woodland, which, according to Bisaz (1968) contain Cembran pine trunks of the ‘ideal shape’ (with solid wood and a continuous axis): the foot of the slope of Staz- and Tais forest near Pontresina exposed to the north. E. Mature woodland with p.min.parte thin stems, 3.4 Growth forms of Pinus cembra multiple trunk non-managed; Fig. 7e, 8 individuals (coexisting types) Areas: National park: beneath Alp la Schera, closed The following growth forms of Pinus cembra are stand 1800-2100 NN; Tamangur near S-char1 found; in each case several individuals are involved: 1200-2200 NN .zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA 185 - Candelabra-shapes (Fig. IOa, b) Candelabra-shapes develop above all when the main stem of a tree has been destroyed by lightening or storm (‘break candelabra’ according to Klein, 1913/ 14), and when earlier lateral branches compete for ‘dominance’ (Mattheck 1991). Particularly bizzare is the appearance of Pinus cembra candelabras, when they have developed to the multiple trunk type and when several main trunks have been destroyed (Fig. 10a, b). The candelabra type of Pinus cembra shown in Fig. 10a was photographed by Klein on 20th August 1987 (Klein 1905, Table 8B) and was rediscovered by us in 1989 (Fig. lob). Klein describes this as ‘the most beautiful Cembran pine (candelabra tree with dominant main trunk) of the Muottas da Celerina (2120 m) . . . height 15-16 m’. Klein (1.c.) records a circumference at breast height of 4,lO m; 1989 we measured 5,05 m. The increase of the trunk circumference by 95 cm in 92 years is considerable (which is an approximate increase of 30 cm in diameter), this can only be explained by the accumulated increment of secondary growth in the thickness of several single trunks and the accumulation of reaction wood. According to calculations made on the yearly increment of single trunk Pinus cembra in the same habitat Klein (1.c.) concluded that the age of this individual corresponds to its five-fold diameter. The candelabra tree photographed by him (Fig. 10a) was, according to his calculation method in 1897, 650 years old. When using this calculation method today, 92 years later, one would arrive at an age of 800 years. This proves that this calculation method cannot be applied to welded multiple trunk individuals, and that other multiple trunk individuals as well are much younger than assumed. Such connections were also not considered by other authors (e.g. Kirchner et al. 1909; Rikli 1909). Formerly, this tree was standing completely free - as can be seen on the photograph by Klein (1.c.) - and it had far more branches and needles; today it is growing in the parklike zone beneath the actual timberline. - Individuals sho wing welding processes which begin at a minor height, appearing to be ‘stilted’in the lower part; in the top area it comes partly to a ‘disentanglement’ All these observations indicate that the rarely occurring stilted forms consist of 2 individual trees, originating from seeds which were deposited far enough apart by the Nutcracker, so that the growing trees had no contact at their bases; neither were they dispersed ornithochorously. Only on a higher level do the trunks come into contact and start growing together. In the crown part the two single tops ‘disentangle’ and can be easily distinguished again from one another. 3.5 Analyses of trunk cross-sections of multiple trunk Pinus cembra individuals Multiple trunk individuals of Pinus cembra possess several cores, which again represent the cores of several individual trunks and do not derive from ramification of branches; this is shown by comparing trunk cross-sections in different heights. In most cases we were able to analyze welded double trunk individuals, however, also examples with 3 cores could be documented. Double trunk Pinus cembra with 2 cores, individual trunks of the same age (Fig. I l a , b) In the Staz forest we found a trunk cross-section of a double trunk Pinus cembra, of which the 2 cores were clearly distinguishable. After analysis of the annual tree rings it was found that the two individual trunks being of the same age came into contact with one another when 14 years old after which they fused. This double-trunk individuals, which reached a diameter of 19 cm, were cut when about 44 years old. The increment up to’the time of contact of the two individual trunks, was 1,5 mm/year; thereafter, for correction purposes, reacting wood developed in order to obtain an homogeneous tensionstate of the trunk surfaces (increment 2,3 mm/year) (Mattheck 1990a). This reacting wood balances the tension (Mattheck 1991, 1992). Since coniferes can only store compression wood, this increment is re- 189 Campell (1950)’ Furrer (1955)’ Holtmeier (e.g. 1965)’ and Mattes (e.g. 1978)’ is the most effective ‘planter’ of Pinus cembra, particularly on the timberline. He chooses ‘safe sites’ for his plantings (not too much snow, not too exposed to wind, and not too much ground frost) complying perfectly with the ecological requirements of the tree species (Aulitzky et af. 1982). The most preferred depots are rocky tops and slopes with markedly structured reliefs (Mattes 1978). His ‘tuft plantings’ could serve as models which are biomechanically very well suited to withstand snow- and windpressure conditions. This ‘ideal’ is also being dealt with in recent literature which recommends the planting of tufts (e.g. Aulitzky et af. 1982 and Schonenberger et al. 1990). Forests, serving as protection against avalanches, should be structured in such a way that groups of trees which are confined to small areas should be the dominating feature (Gand 1983). Pinus cembra tufts and their special growth forms play an important role for the sensitive landscape ecological zones in the subalpine area. A more exact analysis of the growth forms allows the reconstruction of the individual history which owes its specific genesis to the planting by the Nutcracker: a coexistence the roots of which are to be found in the close connection between plant and animal. Acknowledgements We heartily thank our colleague PD Dr. C . Mattheck (Kernforschungsinstitut Karlsruhe/Germany) who drew our attention to the realization of the ‘framework principle’ and who gave us important hints regarding the problems of biomechanics. Prof. Dr. H. Mattes (Miinster/Germany) kindly answered questions concerning the seed collection by the Nutcracker. Many thanks to Prof. Dr. H. Zoller (Basel/Switzerland) and our colleagues from the Botanical Society Basel for a number of stimulating ideas for this work. References Aulitzky, H. and Turner, H. 1982. Bioklimatische Grundlagen einer standortsgemafien Bewirtschaftung des subalpinen Larchen-Arvenwaldes. Mitt. Eidg. Anst. forstl. Versuchswesen 58: 325-577. Birmensdorf. Bisaz, 0. 1968. Die forstlichen Verhaltnisse der Gemeinde Pontresina. Biindnerwald 21: 188-200. Chur. Bormann, F.H. and Graham, B.F. 1959. The occurence of natural root grafting in Eastern white pine, Pinus strobus L., and its ecological implications. Ecology 40: 677-69 1. Durham, NC. Bosshard, H.H. 1974. Holzkunde. Bd. 2. Basel, 312 pp. Campell, E. 1950. Der Tannen- oder NuBhaher und die Arvenverbreitung. Biindnerwald 4: 3-7. Chur. Campell, E. and Trepp, W. 1968. Vegetationskarte des Schweizerischen Nationalparks mit einer Beschreibung der Pflanzengesellschaften von W. Trepp. Ergebnisse der wissenschaftlichen Untersuchungen im schweizerischen Nationalpark 11: 19-42. Liestal. Fischer, F. et al. 1960. Kiinstlich hergestellte Wurzelverwachsungen. Mitt. Schweiz. Inst. forstl. Versuchswesen 36: 15-32. Zurich. Furrer, E. 1955. Probleme und Riickgang der Arve (Pinus cembra) in den Schweizer Alpen. Mitt. Schweiz. Inst. forstl. Versuchswesen 31: 667-705. Zurich. Gand, H. i.d. 1983. Schnee- und lawinenkundliche Grundlagen fur Lawinenschutzwalder. Biindnerwald 36: 99-113. Chur. Glutz v. Blotzheim, U. 1964. Die Brutvogel der Schweiz. 3. edit. Aarau, 648 pp. Graham, B.F. and Bormann, F.H. 1966. Natural root grafts. Bot. Rev. 32: 255-292. New York, NY. Holtmeier, F.K. 1965. Die Waldgrenze im Oberengadin in ihrer physiognomischen und okologischen Differenzierung. Diss. Bonn, 163 pp. Holtmeier, F.K. 1966. Die okologische Funktion des Tannenhahers im Zirben-Larchenwald und an der Waldgrenze des Oberengadins. J. Ornith. 107: 337-345. Berlin. Holtmeier, F.K. 1985. Climatic stress influencing the physiognomy of trees at the polar and mountain timberline. In: Turner, H. and Tranquillini, W. (Eds.): Establishment and tending of subalpine forest: research and management. Proc. 3rd. INFRO workshop. Ber. Eidg. Anst. forstl. Versuchswesen Ber. 270: 31-40. Birmensdorf. Holtmeier, F.K. 1989. Okologie und Geographie der oberen Waldgrenze. Ber. d. Reinh. Tiixen-Ges. 1: 15-45. Gottingen. Kirchner, 0. v., Loew, E. and Schroter, C. 1909. Lebensgeschichte der Bliitenpflanzen Mitteleuropas. Bd. 1, Abt. 1: Pinus cembra: 241-272. Stuttgart. Klein, L. 1905. Charakterbilder mitteleuropaischer Waldbaume I. Vegetationsbilder 2 (1-3, 5-7), Jena. Klein, L. 1908. Bemerkenswerte Baume im GroRherzogtum Baden. Heidelberg, 372 pp. Klein, L. 1913/14. Asthetik der Baumgestalt. (Festrede zur Feier der GroBherzogl. Techn. Hochschule Fridericiana bei Ubergabe des Rektorates) Karlsruhe, 59 pp. Kiibler, H. 1959. Die Ursache der Wachstumsspannungen und die Spannungen quer zur Faserrichtung. Holz als Roh- und Werkstoff 7: 1-9. Kuoch, R. and Amiet, R. 1970. Die Verjiingung im oberen 190 Bereich der Waldgrenze der Alpen. Mitt. Schweiz. Anst. forstl. Versuchswesen 46: 159-328. Birmensdorf. Ludi, W. 1958. Beobachtungen uber die Besiedlung von Gletschervorfeldern in den Schweizeralpen. Flora 146: 386-407. Jena. Mattes, H. 1978. Der Tannenhaher im Engadin. Munstersche Geographische Arbeiten 2, Munster/Westf. 87 pp. Mattes, H. 1982. Die Lebensgemeinschaft von Tannenhaher und Arve. Ber. Eidg. Anst. forstl. Versuchswesen, Birmensdorf, 74 pp. Mattheck, C. 1990a. Why they grow, how they grow - the mechanics of trees. Arboricultural Journal 14: 1-17. Berkhamsted. Mattheck, C. 1990b. Engineering components grow like trees. Materialwissenschaft und Werkstofftechnick 21: 143-168. Mattheck, C. 1991. Trees. The mechanical design. Berlin a.0. 119 pp. Mattheck, C. and Moldenhauer, H. 1990. Intelligent CAOmethod based on biological growth. Fatigue & Fracture of Engineering Mat. & Struct. 13: 41-51. Mattheck, C. 1992. Design in der Natur. Freiburg i. Br. 242 pp. Munch, E. 1937. Entstehungsursachen und Wirkung des Zugund Druckholzes der Baume. Forstl. Wochenschrift Silva 45: 345-352. Rikli, M. 1909. Die Arve in der Schweiz. Ein Beitrag zur Waldgeschichte und Waldwirtschaft der Schweizer Alpen. Neue Denkschrift der Schweizer Naturforschenden Gesellschaft 44. Teil 1 u. 2.; Basel, 455 S., 28 Tafeln, 1 Karte. Schiechtl, H.M. and Stern, R. 1975-1984. Die Zirbe (Pinus cembra L.) in den Ostalpen. Teil I-IV. Angewandte Pfl. soz. 2, 24, 27, 28. Wien. Schonenberger, W. et al. 1990. Okologie und Technik der Aufforstung im Gebirge. Ber. Eidg. Anst. forstl. Versuchswesen 325, Birmensdorf, 58 pp. Schroeter, R. 1908. Das Pflanzenleben der Alpen. Zurich, 806 pp. Sutter, J. 1981. Der Tannenhaher (Nucifraga caryocatactesl Cratschla). Bundnerwald 34: 566-571. Chur. Tranquillini, W. 1979. Physiological ecology of the alpine timberline. Ecol. Studies 31. Berlin ma., 131 pp. Vogt, K.A. and Bloomfield, J. 1991. Tree root turnover and senescence. In: Waisel, Y. et al. Plant roots. The hidden half: 287-306. New York. Wilmanns, O., Bogenrieder, A. and Nakamura, Y . 1985. Vergleichende Studien des Pinus-Krummholzes in den Japanischen und europaischen Alpen. Tuexenia 5: 335-358. Gottingen.