Other formats

    TEI XML file   ePub eBook file  


    mail icontwitter iconBlogspot iconrss icon

Tuatara: Volume 22, Issue 3, February 1977

The Regeneration of the Araucano — Patagonic Nothofagus Species in Relation to Microclimatic Conditions

page 245

The Regeneration of the Araucano — Patagonic Nothofagus Species in Relation to Microclimatic Conditions


In the Southern Andean region there are very distinct forest formations: Reichle (1907), Skottsberg (1910, 1916), Schmithusen (1956), Oberdorfer (1960), Hueck (1966). The diverse forest pattern can be related only partially to soil. Apparently the ecology of these forests is differentiated predominantly by climatic factors. Variability is strongly marked in the forests of South Chile between 37° and 43° S, where forests of Nothofagus are physiogonomically important. Certain species of Nothofagus are much in demand for timber from the natural forests, and Nothofagus may play a very important role in the future if afforestation with native species is begun.

Species of Nothofagus are spread over nearly the whole spectrum of forest associations and its species are, to a certain extent, mutually exclusive. Thus they can be considered differential or characteristic species for particular sites. Starting with this hypothesis an examination has been made of ecological differentiation in relation to temperature of Nothofagus. The final composition of a mature stand of trees has had a life history passing through a youthful phase of selection which is important when reasons are sought for the presence or absence of a species. Measurements were, therefore, carried out only at the level of the lower vegetation, so this research is limited to the thermal requirements and tolerances of Nothofagus saplings and seedlings.

Climatic Assumptions

Traditionally the climate of South Chile is regarded as oceanic and strongly humid. In an ecological sense this can be accepted only with considerable restrictions (Schwabe 1956, Weischet 1970, Weinberger, Romero and Oliva 1972). Recently more differentiated climatic classifications were proposed. Van Husen (1967) distinguishes two zones within the research area of this paper. Between 45° S and 41° S summer rainfall occurs regularly even in the lowlands. This zone extends to the north as a tongue through the mountains (Fig. 5). Between 41° S and 38° S in the Central Valley is the zone of occasional summer drought. Here it is only during the winter that the westerly wind is dominant, while summer weather is more often locally determined. Van Husen (1967) arrives at this conclusion on purely climatic criteria; this conclusion can be confirmed by plant ecology, for field observation interprets the forest diversity as a result of the varied topography of the region. Its very broken relief causes a page 246 small-scale mosaic of local climates, which vary extremely, especially in temperature. Consequently, instead of the concept of a more or less permanent oceanic climate, which would come from ordinary mean climatological data, there is a frequent change of normal weather conditions to the opposite, a continental climate, shown by high local temperature variations. However, average annual temperature ranges indicate an oceanic climate (Fig. 1).

Temperature variations are exceptionally strong under an open sky. Because of the latitude there is intense solar radiation with consequent high temperatures in open sites during daylight but considerable heat loss by radiation at night. During summer, minimum temperatures below freezing point are recorded even at a short distance from the coast. Such a condition must be given special importance, implying a very strong selection pressure. These extreme temperature conditions occur mainly during the summer. This is shown by the annual curves of some parameters that characterise the heat and radiation relationship of a sea coast station (Fig. 1). Inland, especially in the rain shadow of the Coastal Range, the summer is even drier and warmer (Fig. 5).

In this connection the duration of sunshine and total radiation are important. The magnitudes of both influence soil temperatures considerably, and also affect the climate of the air layer close to the soil. In Valdivia on clear summer days solar radiation of up to 400 cal/cm2 are recorded. However, the minimum on rainy days in winter shows only 0.5% of this figure.

As an additional source of information comparison of the annual curves for temperature and intensity of radiation are useful. Heating at ground level is slower than heating of the air by the sun and thus corresponds to general observations. This slowing, however, is not as great as would be expected for an oceanic climate. According to Lundegardh (1957, p. 193) the temperature maximum in an oceanic climate appears only 1 1/2 to 2 months after the highest position of the sun. From this point of view there is a noticeable diminution of oceanic influence in the Chilean case.


Sites, selected for each species, were distributed in the region between 37° 30′ and 42° 30′ S, and were chosen irrespective of whether the forests were closed on open, primary or secondary. The sole criterion for the adoption of a recording site was the existence of spontaneous natural regeneration of Nothofagus.

Temperature and air humidity were recorded for periods of several days (5 to 8 days). All measurements were made in the three warmest months (December, January and February). The measurement periods for each species were distributed during different parts of the summer. page 247
Fig. 1: Annual curves of climatic parameters of thermal and radiation regimes at Isla Teja Station, Valdivia. 1960-1969. Sunshine recorded by Campbell-Stokes heliograph. Temperature from meteorological screen at 2m height. Radiation from Bellani pyrenometer. (Data from Huber 1970.)

Fig. 1: Annual curves of climatic parameters of thermal and radiation regimes at Isla Teja Station, Valdivia. 1960-1969. Sunshine recorded by Campbell-Stokes heliograph. Temperature from meteorological screen at 2m height. Radiation from Bellani pyrenometer. (Data from Huber 1970.)

The equipment consisted of a series of ordinary thermohygrographs. They were installed on small benches in such a way that the measuring elements of the instruments were 25cm above the soil. They were calibrated against a wet and dry bulb thermometer and for additional control a minimum thermometer was installed beside each thermo-hygrograph. Because of topographic difficulty the normal type of meteorological screen for protection against direct radiation was not installed. Instead portable protection against radiation was made out of a folded sheet of galvanised iron, shaped into a tunnel 50 X 50 X 50 cm.

It was installed with the openings on a north-south orientation so that even a low angle of sun was excluded from the shade. Daily extremes and 2-hourly values for temperature and relative air humidity page 248 were read off the daily graphs. This made it possible to determine the following parameters for each day of measurement at every site.

  • The average daily temperature: expressed as the mean of a total of 12 readings (0 to 24 hours); like the next value it typifies the heat requirements. By this measurement were obtained indications of the optima and ecological ranges of a species in north to south, and altitudinal distribution.

  • Mean temperature of the warmer half-day (10 a.m. - 8 p.m.): In connection with other questions the water budget, and stress were of particular interest. Thus the mean temperature of the warmer half-day was determined as well. The comparison of a large number of daily curves showed that air temperature at 10.00 a.m. is generally lower than in the evening at 8 p.m. (Fig. 4). For this reason a calculation was made of the average for the warmer half-day from six temperature readings between 10.00 a.m. and 8.00 p.m. The apparent daily retarding of heating effects is of an artificial nature and due to Chilean Summer Time.

  • Average saturation deficit of the warmer half-day: Atmospheric saturation deficit can be used to characterise evaporation conditions. Temperature, relative humidity and saturation deficit of the air are dependent on each other. With the recorded data it was possible to read corresponding saturation deficit from tables in mm of water vapour tension. Average saturation deficit is a better characteristic than relative atmospheric humidity for defining the evapo-transpiration stress in the water budget of plants (Stocker 1956). The six values, obtained from 10 a.m. till 8 p.m., were also converted into a mean value.

  • Average daily temperature variation: The daily amplitude is suitable to identify the ‘continentality’ of the site. This figure was obtained from the equation:

  • Average daily variation = 2Ma — Mi' — Mi"/2

  • (Ma = daily maximum, Mi' = minimum of the preceding night, Mi" = minimum of the following night.)

  • Average species-specific day curve (Fig. 4): All biotopes in which a species maintains itself show similar temperature climatic features. For this reason the measurements from all sites possessing regeneration of a particular species have been combined and evaluated together. Thus the idea was developed of characterising the ‘ideal site’ with the aid of this species-specific data collection. All corresponding daily curves have been summarised in this way. The results should be interpreted as average values that identify the optimal conditions for natural regeneration. This is equally valid for general heat requirements (temperature summation) as for the corresponding daily amplitudes of temperature.

page 249
Fig. 2: Comparison of summer minimum daily temperatures (—) with normal curve (—) Valdivia. Dec. 1961 - Feb. 1972. 5 cm above ground.

Fig. 2: Comparison of summer minimum daily temperatures (—) with normal curve (—) Valdivia. Dec. 1961 - Feb. 1972. 5 cm above ground.

  • Frequency distribution of minimum temperature values (Table 3): The presentation alone of mean values results in considerable loss of information. In the present case it was valid as a rule that the measurement periods from at least 7 stations were available for each species. Thus it was possible to evaluate the corresponding dispersion of values. The probability with which night temperatures during the warm period of the year drop below defined low limit thresholds was calculated for each ‘ideal site’.

    The experiment assumed that the dispersion was subject to a normal distribution. To test this hypothesis a larger data collection was needed but these data were not available for the minima at 25 cm above ground at any one site. However, for a series of years the lowest temperature at 5cm above soil was measured for Valdivia. (A. Huber pers. comm.). A close functional relationship exists between both magnitudes of parameters. Fig. 2 shows the frequency distribution for these values for 33 summer months. The total number of 993 daily values can be accepted as a representative random sample. The total data collection shows a close approximation to the normal distribution. The top of the curve indicates that apparently a mixture of at least two individual dispersions has been plotted. The study of the reason for this anomaly is a climatic problem not further examined here. It is, however, worth mentioning here the possibility that the assumed individual ranges correspond to different types of extreme weather conditions.

    page 250

    In this connection interest is concentrated on the area of fewer observations on the left sector of the curve, where there is a noticeable deviation from normality. The question arises if this deviation is too serious. Therefore with the aid of the given parameter, the theoretical expected frequency for the occurrence of low temperatures was calculated and compared with the empirical values (Table 1). It must be recalled that by the establishment of species-dependent total collection, an abstract of real sites is accomplished. Each of the samples of data that are combined to give species-specific collections could be integrated from different weather condition-dependent ranges, as is expected for the Valdivia station (Fig. 2). The different sites of any species, however, deviate a little from average values, being partially somewhat warmer or cooler. Consequently the distribution curves for individual sites are shifted in relation to each other. This circumstance leads to a more balanced picture for the ‘ideal site’ after combination of data and thus the results can be accepted as satisfactory.

    Temperature gradients close to the soil are very steep (Geiger 1961). The data for the Valdivia station show the relationship represented in Table 2, showing the advisability also of considering temperatures just above freezing point.

  • Preparation of ecoclimagrams (Figs. 6 and 7): Dispersion of individual data around the mean values has also been considered for the determination of tolerance ranges peculiar to the various species. Derivation of the latter is demonstrated in Fig. 3.

    Every point within the system of co-ordinates corresponds to a pair of individual values. The collection of data belonging to any species results in a distribution of points whose extension still cannot be taken as an expression of the mean ecological range. In the case of Nothofagus obliqua the lowest average daily temperature observed was 8.3° C; for N. pumilio the highest was 11.8° C. These values represent extremes which occur exceptionally during the summer in the corresponding biotopes only. The mean ecological ranges of both N. obliqua and N. pumilio are more restricted and it is for this reason that they are never observed together.

Table 1
Frequency of temperature minima below defined limit values, Valdivia, Dec. 1961 - Feb. 1972. 5cm above soil.
Limit value (0° C) Number of cases in three summer months Expected value Observed value
− 1,0 0,8 0,1
0,0 1,7 1,1
+ 1,0 3,4 3,1
+ 2,0 6,1 6,9
+ 3,0 10,2 12,0
page 251
Fig. 3: Derivation of ecoclimagram for thermal factors from field stations, for Nothofagus obliqua (+) and N. pumilio (o).

Fig. 3: Derivation of ecoclimagram for thermal factors from field stations, for Nothofagus obliqua (+) and N. pumilio (o).

A statistical analysis led to the conclusion that about half of the total dispersion is caused by the normal variability of climatic parameters observed in any given locality. Only the remainder corresponds to local differences. In other words, only one half of total value dispersion characterises the capacity of species to adapt themselves to different biotopes.

In accordance with this assumption the average ecological ranges of species (Fig. 3, broken lines) are obtained by reducing total value dispersions (unbroken lines) to half. The reference points are given by the intersection of corresponding mean value co-ordinates. The resulting graphs are designated as ecoclima-grams.

page 252

Discussion of Results

Nothofagus obliqua (Mirb.) Oerst.

The deciduous ‘roble’ is distributed on the Argentinian side of the mountains only between 36° 50′ and 40° 15′ S (Hueck 1966). In Chile this species advances considerably further north, in its var. macrocarpa to approximately 33° S (Munoz 1968). Here it occurs between 750 and 2,200 m altitude. In the south, dow nto 41° 20′ S the main distribution of this species is in the lowlands.

Fig. 5 shows a schematic transect across the Coast Range near Valdivia. This is the only known region where all species of Nothofagus are together with the exception of N. pumilio. Here Nothofagus obliqua goes up to about 500m on the lower rainfall east face. Generally it can be considered as the dominant species of a tree association which is typical for a region of episodic summer drought. Frequent evergreen companions are Persea lingue Nees and Laurelia sempervirens Tul.

Fig. 4 shows that Nothofagus obliqua is a relatively thermophilic tree. In its ‘ideal biotopes’ the day and night temperatures were higher than in those of all other Nothofagus species. Accordingly N. obliqua is found further into the sclerophyllous woodland where conditions are close to the mediterranean climate areas of California and the Mediterranean. The ecoclimogram in Fig. 6 reflects the tolerance range of this species in respect to north-south (vertical axis) and west-east distribution (horizontal axis). These compass orientations are obviously not to be interpreted in a strictly literal sense for it is not only the geographic co-ordinates which determine the degree of increasing latitude or ‘southerness’ and continentality of a site. These orientations have been transferred to an ecological sense, because localised topography is a deciding factor. In this way flat land will show a stronger continentality than adjacent slopes; the degree of southerness of a slope will be determined by its aspect. Fig. 6 shows a certain continental tendency exhibited by Nothofagus obliqua as does Fig. 4, with a wide temperature amplitude. It has been observed that Nothofagus obliqua never goes down to the west coast, although in breaks in the Coastal Range, near Valdivia, and at Rio Tolten the species grows within a few kilometres of the sea. This species does not penetrate into the coastal belt of Aextoxicon forest. Even in open pioneer communities this light demanding species is absent near the sea. Nor does it grow on the flat foreland of Arauco that extends in the front of the coastal range between 37° 15′ and 38° 15′ S.

In the other ecoclimogram (Fig. 7) there are two parameters (temperature and saturation deficit) that are linked by a positive correlation. The higher are their values then the drier and warmer are the respective sites in summer. It must be recalled that these page 253
Fig. 4: Daily temperature cycle in summer on ‘ideal’ Nothofagus sites. 25cm above soil. n = number of sites for each species.

Fig. 4: Daily temperature cycle in summer on ‘ideal’ Nothofagus sites. 25cm above soil. n = number of sites for each species.

figures are those concerning the climate at the air layer close to the soil. However, these values at best allow indirect conclusions on the edaphic conditions of the site.

Nothofagus obliqua is considered an indicator of fertile sites in the southern part of its distribution, but in the north, N. obliqua occupies sites of shallower soils. Because there are commonly long summer droughts in the hardwood region N. obliqua has to be well adapted to this contingency. In this respect, Fig. 7 shows a certain similarity between N. obliqua and N. antarctica.

The ideal site for N. obliqua, even though relatively warm is not, however, free of frost (Table 3). For south Chilean agriculture this is an important factor because former sites of N. obliqua forest are now intensively farmed. Here summer frosts quite often cause serious damage to wheat during the flowering season.

N. obliqua prefers flat sites but not to the same extent as the next species, N. antarctica. This is true not only for the flat lands of the Central valley. For example one can find N. obliqua in the northern sector of the Coastal Range (Nahuelbuta National Park, 37° 47′ S 73° W) on high plateaus of 1,200m altitude. There its range overlaps with that of Nothofagus antarctica and Araucaria araucana C. Koch. page 254
Fig. 5: Distribution of Nothofagus at different altitudes and expositions of Cordillera Peladar (Coastal Range) 140° 10′S. Rain gauges installed in open sites. At Valdivia situated nearby 136% long term precipitation was recorded.

Fig. 5: Distribution of Nothofagus at different altitudes and expositions of Cordillera Peladar (Coastal Range) 140° 10′S. Rain gauges installed in open sites. At Valdivia situated nearby 136% long term precipitation was recorded.

N. obliqua does not grow on the slopes of the Coastal Range; instead one finds N. procera (syn. N. alpina) with numerous evergreen species. In the lowlands, 1,000 m below, Nothofagus obliqua is again found and has the same type of distribution in the Tolhuaca National Park (38° 12′S 71° 48′W) in the Andes.

Nothofagus antarctica (Forst.) Oerst.

By contrast to Nothofagus obliqua the nirre, which is likewise deciduous, is a slow growing tree or shrub seldom over 12m high. Its distribution in Chile is from 35° 31′S and in Argentina from 38° S, thence southwards to Tierra del Fuego (Urban 1934, Skottsberg 1916). The ‘ideal’ N. antarctica stand is characterised by particularly high daily temperature variation (Fig. 4). On one hand are day temperatures less than 2° below the peak of the thermophilic
Table 2
Correlation of daily temperature minima by measurement at different heights above soil
Temperature minimaat 25 cmFrost (< 0° C) at 5 cm
almost certain
page 255 N. obliqua. On the other hand are mean night temperatures of the same magnitude as of the exclusive Andean and Subantarctic N. pumilio. Nothofagus antarctica is called the ‘beggar’ in woodlands of Patagonia (Kalela, 1941). This species occupies extreme and diverse biotopes; it shows some preference for wet hydromorphic soils as well as for dry volcanic ash. Microclimatic research on studied sites gave evidence that regardless of edaphic differences all sites have the common property of high frost frequency. Fig. 6 shows that N. antarctica is a particularly continental species. On ‘ideal’ N. antarctica sites the temperature drops during 18 summer nights to 0°, or below (Table 3). In the mountains this species occupies the soils of plateaus and high valleys (Fig. 5). Moreover it has a sporadic distribution at the Andean timber-line as a small shrub above the forest of N. pumilio. Apparently cold air streams coming from above and being banked up by the dense, relatively high N. pumilio canopy are the deciding factor. Such conditions are observed for example on the volcano Choshuenco (39° 54′S 72° 02′W) and in the Puyehue National Park (40° 46′ S 72° 10′W) A similar explanation can be obtained from the occurrence of N. antarctica in the Patagonian moorland, where there are basins allowing nocturnal cold air drainage. As well, the temperature characters of the soils there indicate a rather slow increase in daily warming up. This aspect has many parallels in the Northern Hemisphere heathlands and moorlands whose temperature properties have been described by Loetschert (1969).

N. antarctica is also observed in the southern part of the area of distribution of N. obliqua. Only occasionally are both species found together, there being generally a sharp boundary between the two forest types, so that the transition from one to the other may take place within a few paces. In such conditions the areas occupied by N. antarctica commonly are flat valleys with shallow soils, areas called locally ‘nadis’ (Weinberger et al. 1970). Such sites are occupied by ‘Zarzales’, formations which are open and contain a high proportion of thorny shrubs, Berberis, Discaria etc. Embothrium coccineum is frequently present and shows a preference for such sites along with Nothofagus antarctica.

N. antarctica is a notable light demander and in this respect is very similar to N. obliqua. This behaviour is reflected in the high values for both species of the average temperature of the warmest half-day (Fig. 7). Sites studied are frequently open and subject to strong insolation. However, both species show quite different optima and tolerance ranges. This is expressed not only by the data in Table 3, but also in Figs. 6 and 7. This is especially the case for the ranges of high temperatures and saturation deficits; there the ecoclimagrams demonstrate noticeable divergences between the two species. The temperature environment which corrsponds to the cuneiform intercalations between the ranges of the two species (Figs. page 256
Fig. 6: Ecoclimagram for thermal factors showing general heat requirements of Nothofagus spp. and their tendency towards more oceanic or more continental climatic conditions respectively. Temperatures at 25cm above soil for 3 summer months. (See Fig. 7 for legend.)

Fig. 6: Ecoclimagram for thermal factors showing general heat requirements of Nothofagus spp. and their tendency towards more oceanic or more continental climatic conditions respectively. Temperatures at 25cm above soil for 3 summer months. (See Fig. 7 for legend.)

6 and 7), is found predominantly on the eastern flank of the Andes. This zone is specifically occupied by Austrocedrus chilensis and Lomatia hirsuta.
N. antarctica sites are characterised by particularly high atmospheric saturation deficits. There the soils are usually only partially covered by vegetation. They dry out on the surface very rapidly page 257
Fig. 7: Ecoclimagram of Nothofagus spp. for microclimatic factors which influence the water regime. Number of values as for Fig. 4.

Fig. 7: Ecoclimagram of Nothofagus spp. for microclimatic factors which influence the water regime. Number of values as for Fig. 4.

and evaporation decreases to low values. N. antarctica is the only species of the genus which tolerates conditions of the sub-Andean rain shadow steppe formation.

Nothofagus procera (Poepp. et Endl.) Oerst. (Syn. N. alpina Oerst.)

The ‘rauli’ forms almost pure stands as trees up to 40 m in height. It is distributed in mountain areas of Argentina (39° 22′ to 40° 23′S, Hueck 1966), and Chilen (35° 20′ to 40° 20′S). This page 258 summer green tree occupies an intermediate position in its temperature ecology. This does not, however, imply that it will be found together with all the other Nothofagus species.

According to Figs. 4 and 6 the continentality is approximately the same as for N. obliqua. Mixed stands are located in areas of transition between the two species but the ‘ideal’ stand of N. procera is considerably cooler than for N. obliqua. In 46 samples from stands of N. procera, N. obliqua was present in less than one third. A marked plant-sociological link exists with N. dombeyi which was present in two-thirds of the stands sampled. Coincidentally in Fig. 4 is shown a large similarity of the temperature curves of both species (N. procera and N. dombeyi).

N. procera frequently occurs with Lomatia dentata. In general the deciduous N. procera shows considerable affinity to various evergreen trees of the cool oceanic climate. This affinity goes so far that in Valdivia province, on the eastern flanks of the Coastal Range at 800-900 m this species forms mixed species — rich stands with N. nitida (Fig. 5). According to Urban (1934) N. procera appears in the province of Chiloe scattered in the evergreen forests. However, the writer did not succeed in relocating any N. procera there.

According to the ecoclomagrams, the thermal requirements of the sites of N. antarctica and N. procera overlap to a considerable extent. Nevertheless both species are not found growing together because there are other differentiating factors. N. procera is almost exclusively a tree of slope positions, on soils developed in the Andes from recent volcanic ashes. On the geologically older Coastal Ranges it is limited to deeply weathered soils, where competition pressure on these soils is very strong. The excellent ability of N. procera to succeed under such conditions is shown by the fact that it shows good regeneration in heavy shade. By contrast, it has already been emphasised that N. antarctica is light demanding.

N. procera is a species with large and relatively soft leaves, so that more than any other Patagonian species of Nothofagus it resembles northern hemisphere summer green trees. It is dependent on a considerable soil water reserve being available and would be unable to survive on N. antarctica sites.
Table 3
Theoretically expected frequency of summer nights (Dec. - Feb.) with minimum temperatures below given limit values, at 25 cm above soil; for ecological interpretation see also Table 2.
‘Ideal Sites’ for:Temperature Limits
N. antarctica10,917,927,737,949,4
N. betuloidesBecause of lack of data, assumed to be in this position
N. dombeyi0,91,93,97,011,7
N. nitida0,51,53,67,614,3
N. obliqua-1,00,40,92,2
N. procera1,12,65,410,017,0
N. pumilio7,913,320,629,940,5
page 259

Finally, the frequency of frosts is substantially lower in N. procera sites and is a consequence of the temperature regime of slopes occupied by this tree. The interpretation of many daily curves of temperature show that on these sites at night warm air streams flow upwards from the valleys (Fig. 4). Such air flows contribute to a marked lowering of the night cooling effect and may often result in temperature increase. This phenomenon is so marked that it remained even when the average from 62 daily temperature curves was calculated. Undoubtedly this is a mechanism which helps to reduce the number of frosts. This phenomenon is responsible for the fact that the relative humidity of the air of the ‘ideal’ sites for Nothofagus procera is approximately 90% between 4 a.m. and 8 a.m. whereas for all other Chilean Nothofagus, including Nothofagus obliqua, there is a variation in the corresponding mean values, from 95% to almost 100% relative humidity. Consequently dew is scarcer on sites of N. procera.

The occurrence in the same areas of N. procera and N. pumilio is not known to the writer. According to Fig. 6 this would not be expected.

Nothofagus pumilio (Poepp. et Endl.) Krasser

The ‘lenga’, a summer green species reaching 30 m height, forms typical tropophilic forests, according to Schimper's definition. These forests are sharply delimited from all other woodland associations plant-sociologically, and extend with particular uniformity from 35° 30′S in Chile and 36° 56′S in Argentina to the southern end of the continent. In Tierra del Fuego this species grows at sea level (Skottsberg 1916).

N. pumilio and N. antarctica are often mentioned together because of their similar Andean and Subantarctic distribution. However, the ecological requirements of these species are very different as is shown, primarily, by the daily temperature curves (Fig. 4). Where N. pumilio regenerates well no high temperatures are recorded in the midday hours. Regeneration is essentially on shaded sites, where the saturation
Table 4
Climatic data linked with radiation balance, for three summer months. Observation period 1916-1945. ‘Covered sky’: when the mean of 3 daily observations reaches at least 8/10 cloud cover.
Observation StationLonquimayPunta ArenasEvangelistas
Distribution ofN.N.N.
Calm condition as % of 3 daily observations17,010,80,8
Covered sky14,836,974,1
No. of daysNight
Average daily variation (0° C)19,98,13,6
page 260 deficit of the air is maintained within moderate limits. In this respect the limit of tolerance is lower than that of the other deciduous species.

Fig. 4 also shows that the mean night temperatures for both N. pumilio and N. antarctica are very similar. Nevertheless even here there is a characteristic difference in the smaller variability of the temperature minima for N. pumilio sites. Consequently there is a significantly lower frequency of night frost (Table 3). From the fact that the average daily temperature for N. pumilio is generally within a lower range than that of N. antarctica (Fig. 6) it follows that the latter possesses greater plasticity. It is probable that the restriction of N. pumilio in the northern part of its area to the more moderate microclimates of slope sites, is due to the thermal limitation described above.

In this connection it is of interest to compare leaf emergence, which in Nothofagus pumilio occurs up to two weeks earlier than for N. antarctica. N. pumilio flowers and begins to open its leaves when N. antarctica is still leafless. This phenological difference, which was noted by Skottsberg (1916) is generally very marked and also indicates a better adaptation by N. antarctica.

In the southern part of its area N. pumilio also occupies valley flats and formed extensive forests; around Punta Arenas these forests are mainly destroyed by farming activities. This distribution thus raises the question of how this biotope change can be understood. Data on Table 4 and comparison with more northern stations support the following explanation:

Lonquimay at ca.40° S is in an Andean high valley in which the extensive N. antarctica low forest is physiognomically important. N. pumilio is located here only on the higher slopes. The sky is usually cloudless during the warmer part of the year. By contrast the sky of Punta Arenas is overcast at least every third day. Such cloudiness, significantly frequent at Punta Arenas, demonstrates weather conditions which prevent a strong inward or outward radiation. As a result there are considerably smaller daily temperature variations at Punta Arenas.

Moreover, northern sites of N. antarctica owe their thermal characteristics to the circumstance that they are excluded from the equalising influence of the west wind. In the south the effects of such winds are more pronounced, even inland, and for that reason there are extremely pronounced allochthonous climates (Weitschet 1970). The low number of calm days in Punta Arenas, by comparison with Lonquimay, shows that there is a permanent intensive exchange of air masses at Punta Arenas, and the formation of static cold air ponds is less probable. As a result of these conditions there are practically never any night frosts detected during the warm period of the year at Punta Arenas (at 2 m above soil). However, 1700 km to the north, at Lonquimay such frosts on an average occur every tenth summer night.

page 261

Nothofagus pumilio forms the timber line on the Andes more frequently than other species. Above is a dwarf scrub, very rich in Ericaceae. The plant-sociological link of N. pumilio forests to the evergreen forest is slight, but of special importance are Drimys winteri, Maytenus magellanica, Nothofagus betuloides in southern areas and Nothofagus dombeyi that comes from lower altitudes into the Nothofagus pumilio forests in the Andes.

Nothofagus dombeyi (Mirb.) Oerst.

The evergreen ‘coigue’ is distributed in Chile from 35° S to 47° 30′S (Skottsberg 1916) and in Argentina from 38° to 44° S (Dimitri et al. 1950). According to Fig. 4 and 6 it is a tree with its main distribution on sites of oceanic tendency. Growing to a height of up to 45 m, it reaches its optimum development in the high rainfall western slopes of the Andes mountains.

According to Hueck (1966 p. 364) it is a demanding species that needs deep and dominantly moist soils. The present author wishes to modify this opinion. In fact its ecological amplitude is considerably wider, as Nothofagus dombeyi penetrates successfully into the biotopes of all the other species considered in this paper. In the area examined no other species shows a comparable adaptability. In the region of Valdivia, with increasing altitude or on shallow soils of the lowlands there is an impoverishment of the species-rich mixed forest which gives way to almost pure N. dombeyi forest. Its relative insensitivity to cold goes so far that this species goes up to the lower part of the winter snow line zone. For example, in the National Park of Puyuehue at 900 m to 1000 m there are extensive mixed forest with N. pumilio. In the lowlands N. dombeyi penetrates into the areas of frost exposed ‘nadis’ where it is established together with N. antarctica.

The volcanoes Llaima (38° 42′S), Villarrica (39° 23′S) and Osorno (41° 06′S) lie west of the main range of the Andes in a more oceanic climate. N. dombeyi is located here as a pioneer species on lahars (the volcanic debris streams which flowed down from the volcanoes after eruptions had thawed the ice). Lahars represent extreme sites where soil formation is at an initial state. Most species are not able to establish except with the protection of scattered N. dombeyi shrubs. Such sites are essentially free of vegetation and exposed to strong insolation. Therefore high water saturation deficits are frequently observed in the air layer close to the soil. These conditions, however, usually are limited to short daily periods because such localities are open to west winds, normally starting at noon and bringing moist air masses from the Pacific Ocean.

Less frequent in South Chile are weather conditions defined by a warm dry wind locally called ‘puelche’, occurring every summer for some time. This wind advances over the Andes from the eastern grasslands and causes, on open mountain sites, an abrupt temperature page 262
Fig. 8 (upper): Temperature variation near the coast, at Valdivia (10 m alt.) and in the Andes, La Picada (840 m alt.) during an easterly ‘puelche’ wind. 27 February to 2 March 1971. (Lower): Wind direction at Valdivia. At La Picada the wind started only on the morning of the second day (28/2/71).

Fig. 8 (upper): Temperature variation near the coast, at Valdivia (10 m alt.) and in the Andes, La Picada (840 m alt.) during an easterly ‘puelche’ wind. 27 February to 2 March 1971. (Lower): Wind direction at Valdivia. At La Picada the wind started only on the morning of the second day (28/2/71).

increase (Fig. 8). The temperature influence of this wind at the coast is considerably lowered. It is probable that within the region of evergreen forests those ‘Puelche’ weather conditions on edaphically dry sites cause a similarly effective selection pressure as does night radiation in frost hollows.

The La Picada site (Fig. 8) is one of such lahars. This site is on an open mountain saddle. The fact that N. dombeyi, usually considered as a ‘rain forest species’, regenerates well at this site shows that it is exceptionally well adapted. In fact N. dombeyi, in a comparison of 53 evergreen species, had the highest plasmatic drought resistance (Weinberger et al. 1972).

Taking into account the above mentioned results, N. dombeyi can be characterised as follows:

It is a tree with lower tolerance limits to xerothermic conditions than the summer green N. obliqua, N. antarctica and N. procera (Fig. 7). However, N. dombeyi resists considerable atmospheric and physiological saturation deficits, if they are transitory and when the overall climate of the site is oceanic. Within the more continental N. procera belts of the mountains it is generally observed that N. procera prefers the exposed slopes whereas N. dombeyi is found in the wetter more temperate mountain gorges (Fig. 5).

Nothofagus nitida (Phil.) Krasser

This evergreen tree closely resembles the preceding species, N. dombeyi and, its common name is ‘Coigue’ or ‘Coigue de Chiloe’. Its distribution starts from 40° 20′S on the Coastal Range near Valdivia (Fig. 5) southwards to 48° 30′S (Skottsberg 1916). It has not been recorded on the east of the Andes.

Although the daily temperature curves of N. dombeyi and N. nitida are similar (Fig. 4) the climatic limits of N. nitida are page 263 substantially narrower. The ecoclimagram shows that it has a more restricted ecological tolerance than the other species (Fig. 5). The tolerance range is almost completely within the amplitudes covered by N. dombeyi (Figs. 6 and 7). However, it is only occasionally that both species are found in mixed stands as they differ very markedly in their edaphic requirements. N. dombeyi grows well on the wet soils of depressions in the northern part of its area, whereas towards the south a biotope change is observed. On Chiloe Island the wet soils are occupied by extensive N. nitida forests, whilst N. dombeyi is limited to the better drained morainic soils. Further to the south N. nitida is the tree of the flat and permanently wet soils along the coast.

In further comparison of both species the wider tolerance range of N. dombeyi to daily temperature variations (Fig. 6) is essentially based on its adaptation to higher temperature maxima. The dispersion downwards of minimum values is only a little higher for N. dombeyi sites. Finally the average night temperatures for N. nitida are a little below the temperatures for the other species (Fig. 4). Thus there is only a slight difference in the amount of summer night frost (Table 3). As mentioned earlier, even moorland sites in the coastal area can be affected by a certain frost frequency. This applies to West Patagonia where N. nitida occupies flat coastal lands.

The biotopes of N. nitida are defined by permanently high air humidity. Accordingly this tree occupies the high rainfall belts of the montane Cordillera Peladar (part of the Coastal Range) (Fig. 5). In this respect N. nitida is more demanding than all other species considered (Fig. 7), and may be related to the fact that this species has a very shallow root system.

N. nitida shows strong phyto-sociological links with other forests of evergreens, particularly with Drimys winteri, Weinmannia trichosperma, Laurelia philippiana, Amomomyrtus luma and the conifer Saxegothea conspicua. N. nitida is better suited to very temperate climatic sites than N. dombeyi and is present adjacent to one of the few forest formations of south Chile where the forests lack any Nothofagus species. Such are the forests of the western flank of Coastal Range which are characterised by many trees of the Myrtaceae. Southwards of Valdivia such forests increasingly replace Aextoxicon forests (Fig. 5). However, such Nothofagus-absent forests do not penetrate the general area of the following species.

Nothofagus betuloides (Mirb.) Oerst.

The evergreen ‘Ouchpaya’ or ‘Cohue de Magallanes’ dominates coastal areas from 48° southwards. In the Valdivia region it is found only in a few isolated stands located on the highest and most windy position on the Coastal Range (Fig. 5), and inland, on the volcanoes Osorno and Calbuco. This tree does not grow tall and in the main forests of the Magellanic areas few trees are taller than 15 m.

page 264

In south Patagonia Nothofagus betuloides forms the ‘maritime forest boundary’ inland of the oceanic heath and moorland (Skottsberg 1916). Table 4 defines the climatic conditions. Evangelistas is one island on the outer coast where N. betuloides can grow only in the protected hollows. Temperature variations are very small and clear days are scarce on these coasts where there are storm winds at all times of the year. Even in summer wind speeds of over 40 km per hour occur every two days on the average. N. betuloides is well adapted to such an environment, with its compact growth and particularly small leaves.

Unfortunately there are only 11 daily measurements for 2 sites and for this reason it is only possible to estimate its ecological tolerance. Only the average values which could well define the ‘ideal site’ are represented in the ecoclimagrams (Figs. 6 and 7). The position of N. betuloides in these diagrams demonstrates that it is the equivalent in the far south to the two other evergreen species (N. nitida and N. dombeyi). On the other hand it is in the whole of its distribution in contact with stands of N. pumilio which grows on higher sites on mountains and drier sites to the eastward.

In its habit N. betuloides is very similar to stunted examples of N. dombeyi. The climatic tolerance appears similarly wide and certainly greater than that of N. nitida. However, N. betuloides does not grow on such dry sites as Nothofagus dombeyi. In the trans-Andean valley of the Rio Baker (in Aysen Province, approximately 47° 50′ S) N. nitida occurs on the coast; somewhat further inland is N. betuloides, and 60 to 70 km inland from the mouth of the river a zone of N. dombeyi commences. Skottsberg cited this succession as very remarkable. ‘It is surprising to find Nothofagus betuloides replaced by Nothofagus dombeyi inland when the opposite would be expected’ (Skottsberg 1916, p. 71).

However, according to the results of the present study the plant geographic observations of Skottsberg are quite comprehensible. It had already been noted that a warm and very dry east wind (Fig. 8) can penetrate such openings in the mountains as the Rio Baker. The average temperatures between 10 a.m. and 8 p.m. (Fig. 7) indicate that N. betuloides does not tolerate high temperture to the same extent as does N. dombeyi. The distribution in the Cordillera Peladar, where both species are represented, also shows clearly that Nothofagus dombeyi tends towards an inland climate, rather than Nothofagus betuloides.

However, in even more southern and colder regions N. betuloides substitutes completely for the other species. It is there established on edaphically favourable sites as well as to the east of the Andes (Hueck 1966).

page 265


Dimitri, M. J., y Milano, V. A.: Las Plantas cultivadas en la Republica Argentina-Fagaceas. Minist. Agricult. Ganad. (Buenos Aire,) 8, Fasc. 156 (1950).

Geiger, R.: Das Klima der bodennahen Luftschicht. Braundschweig 1961.

Huber, A.: Diez anos de observaciones climatologicas en la estacion Teja-Valdivia (Chile) 1960 his 1969. Valdivia, Universidad Austral de Chile 1970.

Hueck, K.: Die Walder Südamerikas. Stuttgart 1966.

Kalela, E. K.: Uber die Entwicklung der herrschenden Baume in den Bestanden verschiedener Waldtypen Ostpatagoniens. Ann. Acad. scient. fennicae, S.A. Helsinki 4, 1-71 (1941).

Kuhnelt, W.: Grundriss der Okologie. Jena 1965.

Lotschert, W.: Pflanzen an Grenzstandorten. Stuttgart 1969.

Lundegardh, H.: Klima und Boden. Jena 1957.

Munoz Pizarro. C.:. Los bosques de Nothofagus. Conferencia Latinamericana Regional sobre Conservacion de Recursos Naturales Renovables, San Carlos de Bariloche, 2. 4. 1968.

Oberdorfer, E.: Pflanzensoziologische Studien in Chile, Weinheim 1960.

Oficina Meteorologica de Chile (ohne Antorenangabe): Valores normales de 36 Estaciones seleccionadas, Periodo 1916 bis 1945. Climatologia en Chile, Fasc. 1 (ohne Jahresangabe).

Reiche, C.: Grundzuge der Pflanzenverbreitung in Chile (Die Vegetation der Erde VIII). Leipzig 1907.

Schmithusen, J.: Die raumliche Ordnung der chilenischen Vegetation. Bonner geogr. Abh. 17: 1-89 (1956).

Schwabe, G. H.: Die okologischen Jahreszeiten im Klima von Mininco (Chile). Bonner geogr. Abh. 17: 139-182 (1956).

Skottsberg, C.: Ubersicht uber die wichtigsten Pflanzenformationen Sudamerikas s. von 41°, ihre geographische Verbreitung und Beziehungen zum Klima. Kungl. Vetenskapsakademiens Handlingar, 46, Nr. 3, 1-28 (1910).

: Die Vegetationsverhaltnisse langs der Cordilleren de los Andes s. von 41° s. Br. Kungl. Vetenskapsakademiens Handlingar 56, Nr 5, 1-336 (1916).

Stocker, O.: Die Abhangigkeit der Transpiration von den Umweltfaktoren. In: Handb. d. Pflanzenphysiol. III, 436-510. Berlin, Gottingen, Heidelberg 1956.

Urban, O.: Botanica de las Plantas endemicas de Chile. Concepcion 1934.

Van Husen, C.: Klimagliederung in Chile auf der Basis von Haufigkeitsverteil-ungen der Niederschlagssummen. Freiburger Geogr. Hefte 4 (1967).

Weinberger, P., Binsack R.: Zur Entstehung und Verbreitung der Aschenboden in Sudchile. Tropenlandwirt 71, 19-31 (1970).

, Romero, M., and Oliva, M.: Untersuchungen uber die Durreresistenz patagonischer immergruner Geholze. Vegetatio 11, 1-24 (1972).

Weischet, W.: Chile, seine landerkundliche Individualitat und Struktur. Darmstadt 1970.

page 266

The Editor

Dear Sir,

In Table 1 of ‘The Ecology of Nothofagus and Associated Vegetation in South America, (Tuatara 22: 38-68) D. R. McQueen (1976) refers pollen of Nothofagus alessandri to the ‘brassi’ group.

Pollen of the rare N. alessandri is usually placed in the fusca group (Cranwell 1939) but van Steenis (1971. p. 97) raises a doubt when he comments that N. alessandri pollen should be referred to the ‘brassi’ group. an opinion based on a private communication from Dr Archangelsky, Museo de Ciencias Naturales, La Plata, Argentina. Heusser (1971, p. 35, pl. 28-334) describes and illustrates pollen of a specimen, SG 63329, Museo Nacional de Historia Natural, Santiago, identified as N. alessandri which is clearly of the ‘fusca’ type. In a letter (5 April. 1976) in response to a query from me, Heusser states that he has ‘no reason to suspect that N. alessandri represents “brassi” type’ and that therefore his findings are ‘in agreement with Lucy Cranwell's interpretation’. Cranwell (1939) referred pollen of N. alessandri to the ‘fusca’ type, and a duplicate slide of her material in the Botany Division collection supports that conclusion. Van Steenis believes that Cranwell's material, obtained by Skottsberg from a specimen in the herbarium of Arnold Arboretum, was wrongly identified, but this seems unlikely in view of Heusser's (1971) more recent work with a different specimen.

In the present state of our knowledge then it seems that N. alessandri pollen should be assigned to the ‘fusca’ group and that ‘brassi’ type pollen is not now represented in South America, although there is no doubt of its presence in the Tertiary (Cookson and Cranwell 1967). It is important to draw attention to this matter since there is considerable interest attached to the biogeography of Nothofagus in the Southern Hemisphere.

N. T. Moar

Botany Division, D.S.I.R.


Cookson, I., Cranwell, L. M., 1967: Lower Tertiary microplankton, spores and pollen grains from southernmost China. Micropaleontology 13: 204-216.

Cranwell, L. M., 1939: Southern beech pollens. Records Auckland Institute and Museum 2: 175-196.

Heusser, C. J., 1971: ‘Pollen and Spores of Chile.’ University of Arizona Press, Tucson.

McQueen, D. R., 1976: The ecology of Nothofagus and associated vegetation in South America. Tuatara 22: 38-68.

Van Steenis, C. G. G. J., 1971: Nothofagus, key genus of plant geography, in time and space, living and fossil, ecology and phylogeny. Blumea 19: 65-98.