Báo cáo khoa học: "Preliminary dendroecological survey on pedunculate oak (Quercus robur L) stands in Tuscany (Italy)"

Chia sẻ: Nguyễn Minh Thắng | Ngày: | Loại File: PDF | Số trang:10

Thêm vào BST

Báo xấu

52
lượt xem 2
download

Download Vui lòng tải xuống để xem tài liệu đầy đủ

Tuyển tập các báo cáo nghiên cứu về lâm nghiệp được đăng trên tạp chí lâm nghiệp quốc tế đề tài: Preliminary dendroecological survey on pedunculate oak (Quercus robur L) stands in Tuscany (Italy)...

Chủ đề:

Bình luận(0) Đăng nhập để gửi bình luận!

Lưu

Nội dung Text: Báo cáo khoa học: "Preliminary dendroecological survey on pedunculate oak (Quercus robur L) stands in Tuscany (Italy)"

article Original Preliminary dendroecological survey on pedunculate oak (Quercus robur L) stands in Tuscany (Italy) A Santini, A Bottacci, R Gellini Laboratorio di Botanica Forestale, Dipartimento di Biologia vegetale, Università di Fírenze, Piazzale delle Cascine, 28, 50144 Firenze, Italy (Received 22 February 1993; accepted 27 July 1993) paper studies the influence of climate on pedunculate oak radial growth in some Summary — This stands of central Mediterranean Italy. Three populations growing along the course of the River Arno were selected. Core samples were measured, and their growth curves standardized and modellized, in order to isolate the climate signal. The response functions were calculated by orthogonal regres- sion of the variables of the tree ring (dependent variables) and the climate (explicative variables). This paper provides an eco-physiological analysis of the results This study helps us understand how the ecotype of the pedunculate oak has adapted to a Mediterranean climate where water supply is a strong limiting factor. dendroecology / pedunculate oak / water supply / Mediterranean forest / eco-physiology Résumé — Observations dendroécologiques préliminaires sur des peuplements de chênes en Toscane (Italie). Le travail concerne l’étude de l’influence du climat sur la crois- pédonculés sance radiale du chêne pédonculé dans l’Italie centrale. On a choisi et échantillonné 3 peuplements le long de l’Arno. Les échantillons ont été mesurés et les courbes d’accroissement ont été standardi- sées et modélisées dans le but d’isoler le signal climatique. Les fonctions de réponse ont été calcu- lées par régression orthogonalisée entre la variable cerne (variable dépendante) et les variables cli- matiques (variable explicative). Les résultats sont discutés du point de vue écophysiologique. Cette étude aide à comprendre comment des écotypes de chêne pédonculé se sont adaptés au climat méditerranéen où la disponibilité de l’eau est le facteur limitant le plus important. dendroécologie / chêne pédonculé / sécheresse / forêt méditerranéenne / écophysiologie INTRODUCTION from Scandinavia and Russia to the Medi- terranean, from the Atlantic to the Urals and the Caucasus (Gellini, 1985). It is so Pedunculate oak or common English oak ecologically flexible that in certain micro- (Quercus robur L) is the most widespread oak in Europe. Its distribution area extends habitat conditions it can be found in phyto- * This paper is dedicated to the memory of Prof R Gellini, Forest Professor at the University of Flo- improvisely deceased during the preparation of the paper. rence
climatic areas ranging from Lauretum maximum and minimum temper- fall; to mean Picetum. atures. The fact that this species is represented A comparison with data derived from such a vast distribution area can be other studies on pedunculate oak in Italy over explained by its genetic variability. There (Nola, 1988; 1991) has not yielded the ex- pected results. The reason for this is pri- exists a wide number of ecological or climatic types, as well as photoperiodic marily that the environments from which ecotypes, such as the chêne de Juin the samples came were very different and climate is hardly ever the only growth limit- (Q peduncolata var tardissima Simonkaï) which, as the French name suggests, ing factor. The whole combination of the starts its vegetative period very late (Gel- stand’s ecological characteristics (including microclimate) determines the greater or lini, 1985). lesser growth levels (Nola, 1991). It was the existence of such a vast num- ber of ecotypes that prompted our dendro- ecological study of the pedunculate oak MATERIALS AND METHODS population in Tuscany. The aim of our sur- vey is to typify the Tuscan ecotype. 3 populations growing along the We sampled The genus Quercus, which plays a of the River Arno (table I). The first is in course major role in forestry, is currently being the natural park of San Rossore (Pisa), at the studied by many European researchers river’s mouth, where the trees sampled were lo- working on the phenomenon of oak de- cated in 3 distinct subsamples. In these areas cline, or dépérissement du chêne (Young, the soil layer is deep with a sandy texture, and 1965; Petrescu, 1974; Aussenac, 1978; has developed on fluvial sediments that are re- cent but suitable for highly evolved plant forma- Klepac, 1981; Becker and Levy, 1982; tions; these terrains do not present chemical or Bernard, 1982; Ragazzi et al, 1986; Oos- physical limitations except for the fact that they terban et al, 1990). are underwater for a large part of the year due In our study we used the methodology to the emergence of the aquifer. The vegetation proposed by Guiot and coworkers from the consist of a mixed uneven-aged wood of broad- leaves, including pedunculate oak, narrow- Laboratoire de Palynologie and Botanique leafed ash (Fraxinus angustifolia Vahl), Euro- Historique at the University of Saint pean ash (Fraxinus excelsior L), smooth-leaved Jérôme, Marseilles, and the software elm (Ulmus minor Mill), white poplar (Populus package "PPPHALOS" (Guiot, 1990). Ac- alba L) and, sporadically, stone pine (Pinus pin- cordingly, we calculated the response ea L). Pedunculate oak grows on the dominant functions obtained from the orthogonal re- plane. gression analysis of modellized ring-width The second population is in the Cascine Park and monthly climate variables: total rain- in Florence, a former game reserve belonging to
the Grand Dukes of Tuscany; the populations Dendrochronological survey were chosen from the wooded area along the banks of the river. The soil has developed on al- The measurement of the ring widths was done luvial terrain created by the Arno river with peb- at the Silviculture Institute at the University of bles and sandy clay; the result is a deep soil Florence with a CCTRMD (computer-controlled layer with the clay component varying according tree-ring measurement device) connected to to the original catchment basin. The vegetation CATRAS (computer-aided tree-ring analysis consists of a sparse mixed wood, on 2 planes, system, Aniol, 1983, 1987). With this method with a prevalence of broadleaves. Pedunculate yearly increments can be recorded with a resolu- oak is represented by isolated individuals on the tion of up to 0.01 mm. dominant plane. There is no renewal at all of pe- dunculate oak. Using the crossdating procedure, we then checked the validity of the measurements we The third population grows along the Arno in obtained. the stretch between Arezzo and Florence, at Re- Statistical tests (correlation coefficient t(rs) nacci. The soil layer has developed from river (sensu Aniol, 1983) and coincidence coefficient and lake sediments, stratified with clayey sand, (Corona, 1986) and visual comparisons helped clay and occasional pebbles; the result is a us select the series which offered a correlation deep terrain, well drained owing to its sand com- with a t(rs) value at least higher than 3 with the ponent, and suitable for highly evolved forest other series from the same stand. From these formations. The vegetation consists in a mixed we calculated the mean curve for each stand. un-even-aged wood of broadleaves, with pedun- The results in S Rossore, the first area, show culate oak, pubescent oak (Quercus pubescens that the correlations are good or excellent only Willd), hop hornbeam (Ostrya carpinifolia Scop) within the individual substands sampled, while and flowering ash (Fraxinus ornus L). Peduncu- they are very poor (t(rs) < 3) between the differ- late oak grows prevalently on the dominant ent substands and for this reason a mean plane. chronology was not constructed. The second The first 2 populations are part of residual area, Cascine Park, has 7 individual series with plane-growing forests, whereas the third is in a excellent correlations, in terms of both visual hilly area. and statistical comparisons; these series were We sampled dominant or codominant plants: averaged, yielding a mean curve of 168 yr. In 17 at S Rossore; 10 at Cascine; and 11 at Re- the third area, Renacci, 8 curves were aver- nacci. Every sampled plant grew in woods ex- aged, yielding a mean curve that goes from cept for 2 trees of Renacci population that were 1750 to 1989; unfortunately, only one tree dated isolated. We extracted 2 core samples from back to 1750, while the others were much each tree, at a height of 1.3 m from the ground; younger. Statistical and visual comparisons of the samples were taken from opposite sides of the mean curves obtained from these last 2 the trunk, following the direction of the contour sampling areas yield encouraging results: statis- lines on hilly ground, and simply from north to tical comparison show that the 2 curves overlap south in flat areas. We used a 60-cm Pressler for 168 yr; t(rs) 6.13; coincidence coefficient = = increment borer, which was long enough to al- 68.0 (99.9) (fig 1). low us to reach the centre of the tree. Where possible we also took stem disks. Information concerning the stand and the in- Dendroecological survey dividual tree was recorded and later included in a specially designed computerized data bank, using a DBASE III Plus program. This enables In order to obtain as much information as possi- us to store data on each of the trees sampled ble on the behaviour of trees in relation to cli- and to retrieve the information later according to mate from dendrochronological data, these data specific features, for example, data on all the must be processed statistically so as to elimi- trees of a certain stand, or on all those growing nate gradually all information not related to cli- on a certain type of soil, etc. Accurate sampling mate. As a result of this procedure, ring-width is particularly useful in guaranteeing a success- curves tend to lose their usual shape: they be- ful survey (Schweingruber, 1983). come a representation first of the indices calcu-
lieved that this is the period over which rings are lated by standardization by polynomial curves a and, then, of the residues calculated by formed in the northern part of the Mediterranean auto- an regressive moving average (ARMA) modelliza- basin (Tessier, 1984). tion. Finally, the regression between the time The series of yearly growth increases are series of the residues and the climatic param- modellized through an ARMA process (Box and eters is calculated. For this method, refer to Jenkins, 1970), which applies a yearly growth Tessier (1984), Messaoudene (1989), Guiot model to each series. This function expresses (1989; 1990), Brugnoli and Gandolfo (1991), the sum of 3 elements in mathematical terms: the Nola (1991); Santini and Martinelli (1991). climate, which can be expressed as a random function; the trend, meaning the fact that the We used the climatic data provided by the growth of a particular year can be correlated not University of Pisa (Agrarian Studies Faculty) for the period from 1927 to 1988 in our calculation only to the climate of that year but also to the of the response functions for S Rossore; for the growth and climate of the previous years; and cy- cles, or repetitive elements such as parasite in- Cascine and Renacci sites we used data provid- ed by the Ximeniano Observatory in Florence, festations or forest management operations. for the period 1879 to 1988. One way of decoding the climate signal and The climatic parameters we took into consid- distinguishing it from background noise is to use eration were total monthly rainfall (P), mean orthogonal regression with the bootstrap proce- maximum monthly temperatures and mean dure (Efron, 1979; Diaconis and Efron, 1983) minimum monthly temperatures. The monthly and an analysis of the principal components of parameters used cover the 12-month period be- the 24 climate variables considered so as to cal- tween the completion of the ring in year t - 1 culate the relation between the ring-width time and the completion of the ring in year t, that is series and the climate parameters time series the period between October of the year before considered. This relationship is called the re- the formation of the ring and September of the sponse function. This procedure was applied us- ing the software package PPPHALOS (Pro- year of formation of the ring. It is commonly be-
grams in Paleoclimatology: Prevision of Hiatus The pseudo-data set thus established was and Analysis of Linkages between Observation useful for calibration, and the correlation coeffi- and between Series, Guiot, 1990). cients were calculated on this set. These coeffi- cients were used to reconstruct the climatic data Some of the raw tree-ring width series were pertaining to the calibration and verification preliminary indexed with polynomials of different years. The verification years were those that degrees, since the low frequency variance (LFV) were not randomly selected. The correlation caused by long cycles (especially the age of the coefficient was calculated between observed tree) made it impossible to perform an analysis and estimated data, both on calibration and veri- of the high frequency variance (HFV) caused by fication years. Then 50 lots were randomly se- the climate. lected and 50 reconstructions were performed. Every raw tree-ring width series was model- Once the 50 reconstructions have been complet- lized with an ARMA process in an order which ed, a mean correlation coefficient (R) between varies according to the series being considered. estimated and observed data and its relative In calculating the ARMA model to be applied standard deviation (S) were calculated, both for (using GALOTO software, which works by the calibration and verification years. The R val- choosing the best model among different trials ue and the R/S ratio for the verification years with the Akaike information criterion (AIC) an give an estimation of the global significance of autocorrelation of residues at regular intervals the response function (Brugnoli and Gandolfo, was highlighted. This probably indicates that 1991). there are periodicities in the tree’s growth for pe- If the R/S ratio is 1.68, the significance is riods equal to the order of residue autocorrela- 90%, if RIS 2 it is 95%, if R/S = 2.58 it is 99%, = tion and which have a strong influence on and if R/S = 3.33 it is 99.9%; when R/S ratio is growth. These periodicities can be related to greater than 4 the probability of error is less years in which pedunculate oaks had rich or me- than 0.001. Each number is either positive or dium rich crops, as De Philippis and Bernetti negative depending on whether the correlation (1990) also reported, or phytophagous pullula- is direct or inverse. Figure 2 describes the corre- tions, as can be observed on the core that lation; the size of the columns is proportional to sometimes present early wood only. the significance. Positive regression coefficients The response functions were calculated on indicate a direct relationship. This means that the variables of the tree ring (dependent vari- above average monthly climate parameters cor- ables) and of the climate (explicative variables). respond to above-average growth, whereas if The regressors of the explicative variables have the monthly values are below average then the been grouped together in order to make the growth will be below average too. On the other relation more stable (table II). In order to check hand, negative coefficients indicate an inverse the stability and significance of the response correlation. This means that above-average functions, the bootstrap procedure was used. In monthly parameters correspond to below- this kind of procedure the n observations of the average growth, and below-average parameters ring values and the corresponding climatic data lead to above-average growth (Messaoudene, are drawn at random and returned to the batch. 1989).
RESULTS Stand 2: Cascine (Firenze) The results obtained show that (fig 2) trees Stand 1: S Rossore (Pisa) growing in this area need a constant water supply in the period preceding the forma- tion of the ring, especially in October- For the series of residues resulting from February-March. In the ARMA modelling of this sampling area November and March-April the trees also need high tem- we calculated the response functions using peratures, both maximum and minimum, the climate data of Pisa in the years 1927- since late frost has a negative effect. In the 1988. In no case did this function yield sta- period of full growth the only evident direct tistically significant results. Several differ- correlation is with rainfall in the month of ent groupings of the climate regressors June. The trees of this stand had a higher were tried, but none of them ever yielded growth in summer when the rainfall was results that were even barely significant. abundant in the previous months. There The regression was also calculated separ- are no correlations with any of the climate ately for temperature and rainfall, but the regressors in July and August. This means result was the same. that in these months rainfall does not have Our explanation for this lies in the fact any influence on growth. that the pedunculate oaks in S Rossore grow in a habitat characterized by abun- dant groundwater which, in some seasons, Stand 3: Renacci (Arezzo) even surfaces, giving the site an almost marshy appearance (the so-called lame). Because of this, the differences in radial In this stand we noticed that trees in the growth from one season to another are class of diameter can belong to very same minimal. In dendrochronology trees that different age groups. We found several live in this kind of habitat are known as trees aged 50-60 yr growing very close to complacent trees (Fritts, 1976); since their one aged 250 yr. The response functions growth is only marginally influenced by cli- were calculated from 1918 onwards and mate variables, they are not suited to a only for the series which began in that dendroecological study like this one. If we year, in order to ensure that it would be accept this explanation we can also under- possible to calculate and compare the stand why the crossdating between the dif- mean value of the functions. To check ferent sampling plots was so difficult, as whether different results are obtained if the we mentioned earlier. Since the climate period of observation is increased, the re- variables do not determine the high fre- ponse functions for the older trees were quency variance, the oscillations in current calculated starting from 1878. Since in increase are relatively low and growth is these calculations we are considering a characterized by microstand factors, which longer period, the response functions of 2 vary from one point in the site to the next. more trees reached the threshold of statis- In conclusion, we believe that this is a tical significance. This is the only differ- stand in which pedunculate oaks grow in ence we found. In figure 2 we can see that good conditions, at least as far as water the most significant correlation for the mini- supply is concerned, conditions good mum temperature is the inverse correlation enough to compensate for the negative ef- observed in November, July and August; whereas for maximum temperatures, the fect of high temperatures on growth.
and the degree of the damage is directly most significant is the inverse correlation in June. This suggests that the inverse correlated with the needs of the species and the age of the tree, especially when correlation with minimum temperatures in the trees have developed in very moist en- the summer months indicates that high night-time temperatures create unfavour- vironments and suddenly have to face peri- able conditions for the growth of peduncu- ods of drought. Furthermore, the peduncu- late oaks. The inverse correlation with late oak is not equipped with any maximum temperatures in June, when the morphological feature to help it overcome trees are probably in full meristematic ac- periods of drought, as is the case with oth- tivity, suggests that they need low day- er species typical of the Lauretum. time temperatures. The direct correlation between rainfall in the autumn months and growth recorded during the following year can be attributed CONCLUSIONS to 4 possible causes. The first is that, once growth is completed, the bud forms, and it is therefore of crucial importance that there An of the response functions analysis is a sufficient water supply available at that shows, firstly, that there is greater signifi- a time. The second explanation could be that in the response function calculated cance rainfall in October, November and Decem- between the mean minimum temperature ber might favour a new growth spurt, lead- and rainfall on the one hand, and radial ing to the formation of a false tree ring growth on the other: this could mean that (Maugini, 1949), which, however, we did minimum temperatures have a stronger in- not find. The third explanation could be fluence on growth. Secondly, rainfall al- that the roots do not respond to the photo- ways shows a direct correlation with periods governing the evolution of the epi- growth, with the exception of the data for geous part of the tree. With a few excep- rainfall in April in the Cascine stand. This tions, roots will continue to grow if they is due to the marked hygrophilia of the pe- have a sufficient supply of nutrients and dunculate oak. water, until the soil temperatures become In Mediterranean environments the wa- too low (Kozlowski, 1971). As a result, if it supply is the main limiting factor for the ter rains during this period the roots will grow pedunculate oak, which explains why it and when the new growth cycle begins, grows only in areas where the ground wa- the following year, the tree will have a ter can compensate for a shortage of rain- much larger root structure and will there- fall (Bernetti, 1991).This is indeed the fore be in a position to grow more. Lastly, case in S Rossore, where tree growth above-average autumn rainfall levels de- shows no statistically significant response lays phylloptosys which means that the to climate regressors. The pedunculate photosynthesis continues; the result is an oak’s marked hygrophilia has been studied accumulation of carbohydrates which by Becker and Levy (1982, 1983, 1986; helps the growth of the following year. Becker et al, 1986). Who demonstrated The negative effect of minimum temper- that it grows well on acid hydromorphous atures in November, in the Renacci popu- soils where it is practically the only domi- nant, but who also noted that after a year lation, could be explained by the fact that of exceptional drought this tree suffers a this is the period when the buds are terrible decline, so much so that its very formed, as we said earlier. Their dormancy survival is threatened in certain habitats. is induced not only by the short daylight Damage caused by drought is possible period, but also by the low temperatures.
of their dendrochronographer; they would also The direct correlation with the amount of like to thank N Casini, L Sani and M Ciolli for rainfall in February and March appears to their assistance during the field surveys and C be connected to a physiological require- Gandolfo for her useful editorial suggestions. ment. In fact, this direct correlation takes place immediately before the cambium be- comes active again, as Maugini (1949) ob- REFERENCES served in Quercus pubescens Willd and Q ilex and Ciampi (1951) in Q suber L. These authors stressed the fact that the cambium Tree-ring analysis using Ca- Aniol RW (1983) Dendrochronologia 1, 45-53 tras. became active again during a period of equinoctial rains, with high temperatures, Aniol RW (1987) A new device for computer- assisted measurement of tree-ring widths. reaching a maximum of 25°C in April (Mau- Dendrochronologia 5, 135-141 gini, 1949). Aussenac G (1978) La sécheresse de 1976 : in- In order record to above-average fluence des déficits hydriques sur la crois- growth levels the trees in the Renacci pop- des arbres forestiers. Rev For Fran sance ulation need an ample supply of water in 30, 103-114 May, June and July. This is probably due Backer M, Levy G (1982) Le dépérissement du to the fact that the growth pattern of the chêne en forêt de Tronçais. Les causes éco- pedunculate oak is completely predeter- logiques. Ann Sci For 39, 4, 439-444 mined and organized in successive fluxes Becker M, Levy G (1983) Le dépérissement du (Ward, 1964). In fact, within a single chêne. Les causes écologiques. Exemple de growth season the same axis can organize la forêt de Tronçais et premières conclu- several buds consecutively, and each one sions. Rev For Fran 35, 5, 341-356 of them, when its time comes, will start to Levy G (1986) Croissance radiale Becker M, grow. Therefore, an ample water supply comparée de chênes adultes (Quercus robur L and Q petraea (Matt) Liebl) sur sol hydro- during this period is essential to ensure morphe acide : effet du drainage. Acta Oecol, that growth can continue. Oecol plant 7(21), 2, 121-143 These data suggest a further hypothe- Becker M, Levy G, Garreau B (1986) Comporte- sis, which will need to be verified. Since ment experimental de semis de chêne pé- there are still significant correlations be- donculé, Chêne sessile et hêtre en présence tween the growth in the current year and d’une nappe d’eau dans le sol. Ann Sci For the climate parameters for the months of 43 (2), 131-146 July and August, it is likely that the activity Bernard C (1982) Contribution à l’étude du dé- of the cambium does not end in October. It périssement du chêne en forêt de Tronçais. Intervention des champignons pathogènes probably continues, although for exactly des racines. Mémoire de troisième année, how much longer is mainly determined by INRA Clermont-Ferrand, France. 107 p stand and ecological conditions. It would Bernetti G (1991) Appunti sul genere Quercus in be very interesting to carry out parallel stud- Italia. In: Problematiche fitopatologiche del ies on the rhythm of cambium activity, in or- gen. Quercus in Italia (A Ragazzi, R Tiberi, der to offer a more complete interpretation eds). Firenze 19-20 November 1990, 12-36 of the data gathered in the present survey. Box GPP, Jenkins GM (1970) Time series analy- sis: forecasting and control. Hoden-Day, San Francisco, USA, 575 p ACKNOWLEDGMENTS Brugnoli A, Gandolfo C (1991) Analisi dendro- climatiche sull’abete rosso (Picea abies (L) Karst) del Trentino orientale: primi risultati. The authors would like to thank the Silviculture Institute of the University of Florence for the use Monti e Boschi 6, 51-56
Ciampi C (1951) Evoluzione della cerchia legno- Ghobri Scien- These Doctorat Algérie. en en sa in Quercus suber L (dall’ottobre 1946 al ces, Univ Aix-Marseille III settembre 1947). NGBI vol LVIII, 293-304 Nola P (1988) Dendrocronologia della quercia Corona P (1986) Metodologie dendrocronolo- nella pianura padana. Dendrochronologia 6, 179-192 giche, Monti e Boschi, XXXVII, 39-44 Nola P (1991) Primo approccio alla dendrocli- De Philippis A, Bernetti G (1990) Lezioni di selvi- matologia della quercia (Quercus robur L) e coltura speciale. CUSL Firenze 1991, 239 p Quercus petraea (Mattuschka) Liebl) in pia- Diaconis P, Efron B (1983) Méthodes de calculs nura padana (Italia settentrionale). Dendro- statistiques intensifs sur ordinateurs. Pour la chronologia 9, 71-93 Science 46-58 Oosterban A, Nabuurs GJ, Maas GJ (1990) Een Efron B (1979) Bootstrap methods: another look onderzoek naar verbanden tussen sterfte in at the Jacknife. Ann Stat, vol 7, 1-26 zomereikenpstanden in Nederland en de Fritts HC (1976) Tree Rings and Climate. Aca- grondwatertrap. In: Rapport Rijkinstituut voor demic press, London, 567 p Onderzoek inde Bos landschapsbouw "De Dorschamp" nr 532, Wageningen, 60 p Gellini R (1985) Botanica forestale. Vol II. CED- AM Padova, 205 p Petrescu M (1974) Les dépérissements du chêne en Roumanie. Eur J For Path 4, 222- Guiot J (1989) Methods of calibration. In: Meth- 227 ods of Dendrochronology: Applications in the Environmental Sciences (E Cook and L Kai- Ragazzi A, Dellavalle Fedi I, D’Onofrio G (1986) riuskstis, eds) Kluwer Academic Publishers Osservazioni preliminari su un deperimento and IAASA, Dordrecht, 165-178 di Quercus robur L, Quercus cerris L e Quer- cus frainetto Ten in Italia. Atti Giornate Fitop- Guiot J (1990) Methods and Programs of Statis- atologiche 2, 241-252 tics for Paleoclimatology and Paleoecology (J Guiot and L Labeyrie, eds) 253 p Santini A, Martinelli N (1991) Indagine dendro- ecologia sull’abete bianco (Abies alba Mill) di D (1981) Les forêts de chênes Slavo- Klepac en Serra San Bruno (CZ). GBI 6, 895-906 nie. Rev For Fr 33, 86-104 Schweingruber FH (1983) Der Jahrring. Verl P Kozlowski TT (1973) The Shedding of Plant Haupt Bern-Stuttgart, Germany, FR 276 p Parts. Academic Press Inc, New York, 560 p Tessier L (1984) Dendroclimatologie et écologie Maugini E (1949) L’evoluzione della cerchia leg- de Pinus silvestris L et Quercus pubescens nosa in Quercus pubescens Willd e in Quer- Willd dans le Sud-Est de la France. Thèse cus ilex L nel clima di Firenze (dal giugno Doctorat d’État, Université Aix-Marseille III 1946 al giugno 1947). NGBI, vol LVI, 4, 593- 611 Ward WW (1964) Bud distribution and branch- ing in red oak. Bot Gaz 125, 217 p Messaoudene M (1989) Approche dendroclima- tologique et productivité de Quercus afares Young CWT (1965) Death of pedunculate oak Pomel et Quercus canariensis Willd dans les and variations in annual increments related to climate. For Comm For Rec 55, 1-15 massifs forestiers de l’Akfadou et de Beni-