FURTHER OBSERVATIONS OF GEOLOGIC AND GEOMORPHIC CHANGES RESULTING FROM THE CATASTROPHIC EARTHQUAKE OF MAY 1960, IN CHILE

ABSTRACT

A geomorphic study of the region in which the earthquake occurred, made after the earthquakes is compared with a similar study made before the earthquakes. Evidence is presented to show changes in land level. Discussion of the effects of the tsunami as it affected geomorphic and geologic processes is presented. Landslides are listed and analysed. Distribution of in-tensity is discussed.

INTRODUCTION

In May 1960, when southern middle Chile^1 was subjected to the great earthquake, I had just spent one year at the Universidad Austral de Chile in Valdivia as Director of the Geographic Institute. During this time and during a previous residence of one year in 1956, I had completed a series of geomorphic studies in the "little south," and was therefore fortunate enough to have sketches and photographs for many parts of the area prior to the catastrophic changes which occurred. Generous technical and financial support from the abovementioned university and from the German Research Association made it possible for me to study thoroughly the geological-geomorphic and the cultural changes associated with or following the earthquake in the part of Chile lying between 38degree and 43degree south latitude. A preliminary report (Ref. 2) and two special publications (Refs. 3 and 4) covering the studies completed by November 1960 have been issued.

REGIONAL DISTRIBUTION OF MACROSEISMIC EFFECTS OF THE EARTHQUAKE OF 22 MAY 1960, AND EPICENTRAL LOCATION

The U. S. Coast and Geodetic Survey locates the epicenter of the main earthquake of 22 May 1960, occurring at 1512 Chilean time (1112 GMT), at 38degree south 73^(1/2)degree west (Ref. 5). (This would be at a point lying just off the coastline about 85 km north of Puerto Saavedra.) The following observations of macroseismic effect challenge the above epicentral location. Practically no earthquake damage occurred at 38degree south; the first damage is found approximately 2degree further south, in the basin of Valdivia (fig. 1 in the pocket at the back of this issue). Despite its seriousness, this damage is limited to areas within tectonically depressed basins and to areas with special subsurface geological conditions (figure 2). Continuous superficial earth-quake damage begins approximately 100 kilometers south, and a definite center of intensity can be recognized near the northwest corner of Lake Llanquihue, an area whose geological setting is similar to the southern Chilean Central Valley.
^1 In the literature, one usually finds reference to the earthquake in southern Chile. The region affected by the earthquake belongs geographically to middle Chile. The zone from Concepcion to Arauco is still in the south-central zone and the region from Valdivia to Chiloe forms the "little south" of Chile. Geographically, south Chile begins south of Chiloe (Ref. 1).
In the western districts of the city of Valdivia, almost all brick walls were knocked down, steel-reinforced concrete walls were severely cracked, and most of the frame houses were broken or displaced. Takahashi (Ref. 6) estimates that the seismic acceleration for this region was about 25 to 30% of the acceleration of gravity. This indicates a -Mercalli intensity of X. It should also be noted that approximately 20 kilometers east and west of Valdivia, in the towns of Heullelhue and Corral houses with rotten foundational beams survived the earthquake, which indicates that intensity VII of the above scale could not have been reached. These towns, however, are located on the crystalline schists of the coast range, whereas Valdivia is located in the middle of a tectonically depressed basin. All the habitations in the region of the crystalline coastal ranges were practically undamaged, as were the habitationsBasin, after the Land subsidence. In all, some 15,000 hectares of settled and agricultural land in the Vaivia district were submerged near Valdivia. No damage appeared north of the town of San Jose de Mariquina (70 km northeast of Valdivia), whereas to the south, in the farm settlements, many barns and ranch houses were severely damaged. The difference in geological structure on either side of Valdivia caused exposure of upper Pleistocene gravels down-valley from San Jose de MIariquina. These gravels are found 40 meters below the level of the river in the basin of Valdivia, and are overlain by sandy and clayey alluvial sediments.
In the region of the southern Chilean Central Valley, located between the crystalline coast range and the high Cordillera, the direct results of the earthquake become more obvious as one goes south from the Province of Valdivia. The limited damage to buildings indicates an intensity of VII for the cities of La Union and Rio Bueno and their environs. In Osorno, there are few damaged houses, except for those deformed by failure of the river bank. The intensity here must be calculated as VIII, since almost all masonry walls and chimneys fell down. South of the city, damage rapidly becomes greater. In Rio Negro and environs, numerous houses collapsed, 2 and in the region between Frutillar and Octay, near the northwest corner of Lake Llanquihue, hardly a grain or hay shed was left standing or a house undamaged. North of Octay many cement walls fell down, and in doubly locked, solid, wooden houses, doors and windows were opened. In other places, 30- by 40-centimeter window panes were splintered without damage to the window frames. Between Osorno and Frutillar the railroad was disrupted at no less than 50 places, without particular subsurface phenomena being apparent at the surface. In Llanquihue, the reinforced concrete highway bridge was compressed so that water pipes along its side were bowed up. These phenomena justify the assignment of intensity X. The subsurface here is composed of exactly the same material as it is north and south of this region. Near the lakes there are, in part, well-cemented young moraines, and, farther away, gravel beds of the late Pleistocene age, situated similarly to those of the subsurface around Osorno, La Union, and Puerto Montt.
Damage south of this area becomes less severe. Excluding in this overall view the especially poor building sites in some parts of Puerto Montt (in such parts, according to Takahashi, an intensity of XI was reached), a maximum of VIII occurred in the parts of the town situated on solid ground (well-cemented moraines or gravel terraces).
The rapid decline of seismic intensity, as one goes from the central valley into the Cordillera, is interesting. Consider the following observations in a west-to-east profile just north of Lake Llanquihue: From the northwest corner of the lake where the intensity of X occurred as described above, to La Cascada on the northeast side of the lake, fallen wooden houses occurred (intensity IX). At Ensenada, at the east end of the lake, books fell out of shelves and chimneys fell from the roofs. A little farther east, on the west shore of Lake Todos los Santos, high stacks of dinner plates in the Hotel Petrohue remained standing, and at the Fundo Puntiagudo, as could be deduced from objects placed in unstable positions, the intensity could only have reached a maximum of V or VI. Here in the region of the reactivated north-south juncture of the Cordillera, the main shock was obviously restricted to a narrow zone along active faults, as will be fully discussed below.
The overall view shows an elliptical arrangement of isoseismic lines in the region of the central valley with a center near the northwest corner of Lake Llanquihue (41ーS) and an isolated center in the Valdivia Basin.
Should not this state of affairs point out that the seismic determination of the epicentral location is the location of the weak beginning of the quake, which focused in a north-south trending line, and that after an initial release of energy in the northern part, the main energy was released further south?

THE EPIROGENETIC AND TECTOGENETIC MOVEMENTS, THEIR EXTENT AND INTENSITY

During the first hours following the earthquake of 22 May, all meadows along the lower reaches of the rivers between Rio Imperial and the south of Chiloe were
^2 Translator's note: See Thomas, et al, on Rio Negro.flooded. In the basin of Valdivia at about 1700 hours, I observed the waters in the Calle-Calle River flowing upstream with a greater velocity than I had ever seen the water flowing downstream. Days later, at the paper mill about 30 kilometers up-stream from the river's mouth, marine fishes and salt water were observed. During the following month, water flooding the meadows became sweet. There occurred a loss of meadows and farmland through flooding, which amounted to 15,000 hectares in Valdivia alone.
One can ascribe this, on the one hand, to compaction of the sandy and clayey sediments, which are 40 meters thick in the Basin of Valdivia, and, on the other hand, to a definite epirogenetic-crustal movement. Because of an absence of a fixed reference for leveling, it was difficult to distinguish between the two geologicprocesses. The distinction was also difficult because local subsurface conditions were unknown prior to the earthquake and only the results of the earthquake could be studied. Those already familiar with the basin had no doubts from the beginning that the major process was one of crustal deformation (Ref. 3).
Investigations already in progress showed that an epirogenetic depression of large areal proportions, associated with the earthquake, bent downward a strip of the continent about 30 kilometers wide and approximately 600 kilometers long. Along the coast, depression amounted to 1 to 11 meters, thereby causing the marine transgression. As evidence, the following observations from various parts of the affected continental strip can be cited:
Definite evidence of crustal deformation is seen in documented photographs from the Bay of Mehuin. The Universidad Austral established a station for marinebiology there in the years prior to the earthquake. In front of the station was a fossil-abrasion platform eroded in vertical-standing phylites and crystalline schists on which zones of littoral fauna were established, as could be definitely seen in a photograph. From the photograph, it could be definitely established that water during normal tidal variations did not reach the upper levels of the platform. Today, the platform is constantly under water. An estimate of the amount of depression is possible since the high tide reaches a drainage ditch of the former station which formerly was 1ス meters above high tide.
A comparison of figures 3 and 4 shows the sinking of the land. They show the former and. present conditions near the area of the Marine Biology Station.
Further evidence for crustal movement is the fact that northwestward from the mouth of Rio Tolten, a cliff was cut into the dune sand of the almost flat coastal area. In places, the cliff cuts across a former cattle and wagon trail, as can be seen in figure 5. Geomorphic considerations show that a cliff in a flat coastal area can only be a transitional feature of short duration. In time, the equilibrium conditions of a flat coastal area will again be established.
Some fairly exact estimates of the magnitude of submergence could be made from geomorphic criteria at places along the coast. Such was the case in Puerto Saavedra, at the mouth of the Rio Imperial (fig. 6), where the difference between the new high water level in the foreground and the lowest pine-root level in the background is 1.35 meters. The true continental submergence is probably 1.5 meters, since the high water level must never have reached the lowest root level. The new high water level is at an exhumed bed of concretions.
East of Chiloe, the near shore vegetation has been dried up under the influence of salt water encroachment. At Quellon, on the east coast of Chiloe-about 70 km south of Castro, the minimum high water level is 1.30 meters above the roots of near shore bushes.
On the Rio Maullin, the depressed coastal strip extends upstream to Puerto Toledo, on the Rio Bueno almost to La Union, on the Rio Valdivia-Calle-Calle to Huellelhue, 20 km east of Valdivia, and on the Rio Cruces to San Jose de la Mariduina. The harbor of Puerto Montt, which is farther from the shore, was not depressed.
The northern end of epirogenetic depression must be on the southern part of the Arauco Peninsula (150 km north of Puerto Saavedra), where St. Amand (Ref. 8) and Watanabe and Karzulovie (Ref. 9) report an uplift of 1.2 meters. In the south, the Hydrographie Service of the Chilean Navy has determined a 1 to 3 meter uplift of Cuafo Island (southwest of Chiloe), where two formerly separated islands are now connected by a strip of land. Therefore, the southern limit of depression could be at the road, south of Guafo toward Chiloe.
In summary, there has been a submergence of the Chilean coast from the southern Arauco Peninsula to southern Chiloe. The depressed coastal zone is about 30 kilometers wide on land, and the coast is now around 1.5 meters lower. Compaction of alluvial sediments amounted locally to a few decimeters and was governed by local conditions.
^3 Translator's Note: See Seviers on the Tsunami (Ref. 7).
Of considerable geologic interest is the fact that, after submergence, the coast was at an older level, as shown by geomorphic features. The formerly fossil abrasion level in phylites at ddehuin has now been lowered into the region of surf action. Alpo mentioned was the exhumed plane of concretions at high water level near Puert Saavedra.
Figure 7, depleting an old surf-abraded cavity, shows that the new coastal level was occupied in the past. This old cavity has been covered with green algae in the months following the earthquake. Behind the isolated cliffs from which the picture was taken, there is a series of hollows excavated by the surf, which served as storage places for fuel prior to 22 May, and into which tile surf has again returned at high tide.
It must be assumed that a submergence to an older level of a formerly emergent coast has occurred through crustal deformation.

THE SEISMIC SEA WAVES AND THEIR EFFECTS

Although there are no direct natural observations, one can reasonably assume that the continental shelf was included in the crustal deformation. Five seismic sea waves were recorded by the tide gages of the Chilean -Navy at Valparaiso (Ref. 3). The first wave reached the coast at 1530 Chilean time, the first destructive breaker at 1620, and the last at 2130. A description of the maremoto is given in Ref. 3. Figures 8 and 9 show the lowering of the sea level in the Bay of Corral prior to the second wave, and show the wave itself, which was estimated to be about 8 meters high. The height of the waves varied with the configuration of the coast and the coastal bathymetry. An absolute height of 11^(1/2) meters was determined at the southcorner of the Bay of Puerto Saavedra. Wave heights measured 8 meters in 1lehuin, 8 meters in Corral, 7^(1/2) meters in Muiculpue, near Bahia Manza, and 5 meters in Ancud.
The radical changes which occurred in the Bay of Queule, Province of Valdivia, can be seen in figs. 10 and 11, which were taken from the same place before and after the catastrophe. The breakers were still 4^(1/2) meters high after surmounting the zone of dunes in front of the town, and the town was washed away. It is interesting to note, in the figures, the formation of an estuary from the formerly meandering creek, caused by the change in sea level.
In Puerto Saavedra, only a field of ruins remains of the former heavily populated settlement. The remains of houses are found 3 kilometers inland at the foot of the coastal hills. Houses at the shore were transported with their foundations to the highest area reached by the water, and the tides have changed the canals to 3-meter-wide rills. The whole town of Puerto Saavedra is probably a loss.
In Ancud (fig. 12), the strip of damage can be plainly seen. Here, the foundations of the fishing settlement can also be seen in the highest area reached by the water.
Aside from the destruction of coastal settlements, the submergence of the coast and the seismic sea waves have led to slower physiographic changes. The location of the mouth of Rio Imperial has been basically changed (fig. 13). After joining withthe Rio Moncul, it formerly flowed 4 kilometers southward, before emptying into the sea, along a stretch where coastal sand dunes formed a relatively wide embankment. The northern crown of this dunal embankment was 30 meters high, and some remains of possible interglacial gravels and tuffs occur here. The southern area of dunal embankment was 15 to 20 meters high; the middle area was only 5 to 8 meters high and about 500 meters long. The littoral area on both sides of the embankment was bounded by a flat, sandy shore. The seismic sea waves broke over the saddle and transported some material into the adjoining marshy meadows, and formed a dam across the stream channel. This resulted in a geomorphically unstable situation. Where the river now crosses the dunal embankment, it is very shallow, and the surf breaks across it. In time, this channel will probably be closed when deposition by the surf exceeds the transporting power of the stream, such as during the time between the low summer river level and winter flood. The river will then probably return to its former deeper channel. At present only a portion of the dunal embankment is left, and it is connected to the land by a sandy embankment. The relatively higher water level has resulted in the formation of a small scarp in the sand, which at high tide is undercut by the surf, and which collapses during periods of ebb tide. In time, the shore will again become flat and the dunal embankment will remain significantly smaller than it was prior to coastal submergence.
The deepening of the Bay of Corral and the lower reaches of the Rio Valdivia is of geomorphic and commercial interest. Prior to the earthquake, the river and the bay were somewhat sanded in because of hydraulic gold mining in the middle reaches of the Rio Cruces. Ocean-going ships had difficulty going to Corral, especially at the sand bar Las Tres Hermanas, which was spreading toward the navigational channel and was at most only 1^(1/2) to 2 meters below the water. After the earthquake, new soundings by the Hycirographic Service of the Chilean Navy showed that at the critical point of the channel, between the northern part of the sand bank and Punta Laurel, it is now 5 to 6 meters deeper, and that the bar is now 3^(1/2) to 4 meters below the water. If coastal submergence is assumed to be about 2 meters, an erosional deepening of the channel and the water over the bank amounts to 1^(1/2), and 2 meters, respectively. The end effect is that not only Corral but also Valdivia, 20 kilometers up-river, can be reached without difficulty by ocean-going vessels.
Eroded material was not only transported seaward but also up-river, as shown by cobbles of vesicular basalt deposited in low places along the river. Because of their low specific gravity, these cobbles were transported up to Valdivia. The greatest amount of inland-transported material was not deposited along the river because it runs perpendicular to the direction from which the sea wave came; the material was deposited behind the Bay of San Juan.
In the Straits of Chacao, between the continent and Chiloe, the erosional effects of backwash, which occurred between the sea waves, is evident. The straits are bounded by sea cliffs cut into glacial and glacio-fluviatile sediments. Normally, currents in the straits are navigated with varying success by fishing ships. In many places, erosional indentations were cleaned out by tsunami waves and their strong backwash (figure 14). The material transported from the indentations is found in new bar-like deposits along flatter portions of the shore.
Where the seismic sea waves crossed the Bay of Ancud and impinged squarely against the steep coastal slopes, the cliffs retreated visibly (figure 15). In some places the surf action removed 10 meters of the Pleistocene sediments, and in others a combination of surf and slump, or caving, accounted for a removal of 40 to 50 meters.
In Golfo Ancud and Seno Reloncavi and to the east of Chiloe, no change in the configuration of the coast by erosion or surf can be seen.

LANDSLIDES, SLUMP, AND THE FORMATION OF CRACKS IN AREAS OF MAJOR REACTIVATED TECTONIC TRENDS

In discussing this complex situation it is certainly useful to clarify the state of the situation prior to the event. In the area of interest, there are no geologic maps with a scale greater than 1:500,000. The most recent general geologic map of Chile (Ref. 10), at a scale of 1:1,000,000, shows no structural elements and only the most general geologic units. This situation probably exists because the whole Cordillera in the "little south" has a thick forest growing on a varying thickness of recent tuff. Therefore, mass transport created many new outcrops.
Most landslides and slumps are concentrated in a relatively narrow strip which, surprisingly enough, is not at the foot of, but within, the high Cordillera. When one flies over the glacial furrow occupied by the southern Chilean lakes, into the Cordillera, the first sparse, local landslides appear 20 to 25 kilometers from the mountain range's topographic boundary. At a distance of 25-30 kilometers from this boundary, a definite zone appears in which the slides are close together and mountain flanks are bare for many square kilometers because of a slumping of the tuff layer or landsliding of weathered rock. In many parts of this zone only 1/4 of the forest remains. The zone ends 8 to 10 kilometers to the east as definitely as it begins. This zone is about 10 kilometers wide in the south at Lake Todos los Santos, 8 kilometers wide at Lake Rupanco, and 10 kilometers wide at Lakes Ranco and Panguipulli. From the east end of Lake Calafquen, to the Pellaifa, hinterland, the zone broadens to almost 20 kilometers, and shrinks to 8 kilometers again at Volcan Villarica. Further north, no important landslides were encountered. In the zone, not only have sections of the tuffaceous cover slid over the underlying rocks, but landslides in volcanic and plutonic rocks have also occurred. The former occur in the Sierra, Santa Domingo, and Cerro La Picada on both sides of Lake Todos los Santos, and the latter are especially well seen on both sides of the eastern end of Lake Rupanco at Puntiagudo.
Once the topographic conditions for some of the slides were recognized, a likely zone for the localization of the slides was suspected from the topography. The suspected zone trends N13degreeE in a line beginning at the north end of the Estero Reloncavi, extending across Ensenada Cahutue, over the volcanics of Puntiagudo, along the eastward extension of Lake Rupanco, and from the group of volcanos around Casablanca to those around Puyehue where the eruption on the day following the earthquake occurred. Further to the north, the line extends just east of Lake Ranco, across the furrow of Lake leltulme (east of Lake Panguipulli), over the divide between Lakes Calafquen and Pellaifa to the volcano Villarica. The suspected line was confirmed by field observations. The zone of landslides points out a zone of depression within the Cordillera, which is broken by younger volcanoes, as can be plainly seen from the air. For instance, in the valley north of the volcano Villarica (figure 16), the westerly wall is gentle and the easterly wall is cliff-like. This asymmetry is marked by the fresh landslips, which have removed much of the vegetation and weathered cover on steeper slopes, exposing light-colored silicic plutonic rocks, granite or granodiorite. The rocks themselves tell us little about thestructure, but the land form is one of tilted blocks that have been uplifted and faulted more on the east than on the nest. The areas between these tilted (uplifted) blocks are not the types of valleys seen elsewhere in the Cordillera. They are probably not erosional valleys formed by ice, since there is not sufficient area for ice accumulation, and there are no traces of glacial abrasion on the valley walls. Instead, there occur outcrops of landslip debris up to 50 meters above the stream in contact with volcanic ash upstream. These deposits form terraces in the outer reaches of the valley. From these observations, it can be concluded that the areas between the tilted blocks were filled with pyroclastie sediments.
Similar conditions occur on the north side of Volcan Puyehue. Figure 17 is a view south toward the volcano, showing the elongate trough. In the foreground, along thetectonic trend, is the Maar de Caran and the crater of Rininahue. Both have been formed in recent decades. In the southward extension of the valley are found the lava flows extruded from the seven vents active after the earthquake.
The eastern boundary of the fault zone is again in plutonic rocks, which, because of their scaly structure, promote the formation of steep mountains. In these plutonics the zone of landslip ends as suddenly as it begins. Farther south, in the area east of Lake Rupanco, similar conditions occur.
From the above observations it can he concluded that the zone of landslip marks a graben-like structural zone with tilted blocks located within the Cordillera. This fault zone was shaken by the earthquake, but no slip along the fault during the quake could be shown.^4
^4 Translator's note: This zone is the trough of the Reloncavi fault zone.
A question still remaining concerns the cause of the landslip zones in the area east of Lakes Rinihue, Panguipulli, and Calafquen, which are sometimes twice as wide as those further south. In the cross valleys of Lakes Calafquen and Pellaifa, there occur deposits of landslip debris in a 22-kilometer-long, east- and west-trending body. This does not include a second 1-to-2-kilometer-wide zone of landslip just west of the above-mentioned lakes.
The following observations were made around Lakes Calafquen and Pellaifa (Figure 18). On the south side of Lake Calafquen, the unrilled, little-eroded wall is covered with landslide scars. These scars expose outcrops of gently westward-dipping volcanics. A large rockflow of landslide debris has come from the first cross valley,that of Estero Lingoico, and it has entered the western part of Lake Pellaifa, damming it so that the water became about 8 meters deeper. The Lingoico cross valley marks the contact between volcanics on the west and granodiorite on the east. The contact can be followed along a N13ーE strike to the south side of Lake Villarica and from there to a similar small graben. It is therefore an extension of the aforementioned fault zone.
The landslips are not confined to the aforementioned places. Eleven kilometers eastward, in the granodiorite, slips occur in greater profusion.
One of the largest landslides occurs on the south side of Cerro Aniques, west of Lake Calafquen, and its deposits obscured the Hot Springs of Malihue, including inhabitants and cattle. On the flank of Aniques, a series of gray lamprophyre dikes striking SSW-ENE occur, which are morphologically expressed as slots. Theybelong to the tectonic feature shown in figure 1, the line lying west of Lakes Rinihue, Panquipulli, and Calafquen. This tectonic feature is a graben. During the earthquake the walls of this graben were severely shaken, and on the west side of Lake Rinihue this led to the greatest deposition of rock debris (Ref. 4). In the steep-sided canyon of the Rio San Pedro, which is 80 meters deep and has a maximum width of 500 meters, 4 landslip dams occurred. The total closure of these 4 dams is estimated to have been about 5 million cubic meters. The largest of these (figure 19), which originated from the river bank, was more than 1,100 meters wide, and the lowest part of the crown was 40 meters high, i.e. 26 meters above the level of Lake Rinihue.
This large landslip deposit traveled 500 meters horizontally with a vertical drop of only 50 meters. This was possible because of special petrographic and stratigraphic conditions. The Rio San Pedro was first cut into 80 meters of Pleistocene sediments, of which the lower 40 meters are finegrained, thixotropic, varved lake sediments. Upon being vibrated, these sediments practically liquefy, becoming a porridge-like flowing mass. This occurred during the earthquake, and the upper 40 meters of conglomerate, sand, and morainal material slid across the canyon over the finer, liquefied sediments. All available methods of lowering the dam were tried, but in the two months of attempts to lower the spillway, the weather succeeded in filling the lake 26 meters above its former level, and about 3 million cubic meters of water stood ready to break through the dam on 24 July when overflowing began. In 30 hours, the largest dam was completely cut through. Fortunately, the river entered a 50-meter-wide winding canyon just a few hundred meters downstream from the dam. In spite of this canyon, the 1-kilometer-wide floor of the valley on either side of Rio Calle-Calle was flooded d and 6 meters deep just east of Valdivia, between Los Lagos and Cuesta (figure 20). As a result of this flood, many buildings were carried away, areas in the valley were covered with gravel and sand, hundreds of thousands of cubic meters of wood were carried into the sea, and the railroad grades and roads were washed away. Because of adequate warning, no people were injured.
Westt of the intramontaine graben and the Lakes Rinihue, Panguipulli, and Calafquen, the SSW-ENE trend is conformable with the Andean foreland. The interruption of the central Chilean valley by the coastal Cordillera occurs along this trend, as do longer stretches of the coast north of Valdivia.
At the northern end of the SSW-ENE-trending coastline between Corral and Mehuin, where it turns northward, there occurs a crack formed during the earthquake (figure 21). This crack cuts the E-W-striking structure of the crystalline schist in a SSW-ENE direction and can be followed hundreds of meters up the coastal embankment until it disappears under soil and forest. In solid cliffs, the crack varies in width from 1 to 2 centimeters, and it shows no differential movement.
With moderate caution, let us consider observations of landslip in the flatland west of Lake Llanquihue. Noteworthy are the numerous landslides at isolated places on the steep banks of the Maullin River by Puerto Toledo, even though no special conditions of the bank or its material at these places exist. Possibly these are the slides which Tazieff called attention to (Ref. 11), and which Lomnitz also observed. It is worthy of note that these slides are located on the projection of a lineament marked by geologic and geomorphic features. The lineament begins atthe volcanics of Puntiagudo, extends over a ridge to the volcano of Osorno and its row of fresh parasitic cones accompanied on both sides of the Cordillera La Picada by many landslides. The lineament crosses the volcano Osorno, on whose northern side slides in volcanic ashes occurred during the earthquake, destroying a ski hut. Along the lineament's extension on the southwest side of the volcano occur the youngest parasitic cones, and eastward from the foot of the Bahia del Volcan landslides in basalt occur in a confined zone. The lineament extends further along the peculiarly straight SE shore of Lake Llanquihue, and from there out into the flat land to the banks of the Mlaullin River where the above-mentioned slides occurred. The river bed itself is peculiar at this place in terms of its hydrology and geomorphology. Along certain stretches, the river is hidden from the aerial observer by a dense heath, and the water emerges from the heath in two relatively low waterfalls. Time did not allow a more thorough investigation of details. However, all my observations from Puntiagudo to the above-mentioned waterfalls support the idea that this lineament may have been formed as a manifestation of movement in the Andes similar to that studied in the Mexican Cordillera.
In conclusion, consider some observations on the number of earthquakes similar to those of 22 May 1960. In front of the unstable upper cliff of Cerro Aniques, west of Lake Calafquen, about 150 meters above the valley floor, there begins a heap of landslide debris. The center of this heap has been cut to its bottom by a creek. Exposed in the creek walls, the granodiorite basement is first covered by 2 meters of gravel mixed with lapilli ash. Then follows a light colored lapilli tuff. Above this, there occurs only landslide debris cut at four horizons by humus (soil).
In the whole Cordillera of Southern Chile one can observe that the fresh, ice-polished basement is covered by a tuff without the development of a weathered zone. One knows from the sequence of tuffs overlying the moraines of the Andean foreland that the tuff was deposited immediately after the deglaciation (Ref. 12).
Therefore, if four landslides occur over the tuff at the foot of Cerro Aniques, one can assume with certainty that, since the last glaciation, or in the last 6,000-8,000 years, this tectonic lineament has experienced four earthquakes of a magnitude and with extraordinary results like that of 22 May 1960.

REFERENCES

1. Weischet, W. Chile. Landeskundliche Ubersieht nach Beobachtungen auf einer Forschungs-
reise im Jahre 1956. Mitt. Geogr. Ges. Muchen Bd. 43 (1958) S. 103-140.
2. Weisehet, W. Die geographischen Auswirkungen des Erdbebens vom 22. Mai 1960 im
Kleinen Suden Chiles (vorliiufiger Bericht). Erdkunde XIV (1960) S. 273-288.
3. Weischet, W. Contribuciones al estudio de las transformaciones geograficas en la parte
septentrional del Sur de Chile por efecto del sismo del 22 de mayo 1960. Univ. de Chile,
Inst, de Geol., Publ. Nr. 15, Santiago 1960.
4. Weischet, W. Voraussetzungen, Vorgang und Folgen tyxotroper illassenverlagerungen ins
Tal des Rio San Pedro (Prov. Valdivia/Chile). Mitt. Geogr. Ges. dluchen 45 (1960) S.
39-50.
5. Rothe, J. P. Le tragique bilan des seismes de 1960. La Nature 1960. S. 378-387.
6. Takahashi, R.Informe preliminar. Prensa de Valdivia. Julio 1960.
7. Seviers, C., H. A. Maremolo of 22 May 1960.
8. St. Amand, P. South Chile Earthquake Swarm of 1960. Geol. Soc. Am., Annual Meetings
1960.
9. Watanabe, T., u. Karzulovie, J. Los movimientos sismicos del mes de mayo de 1960 en
Chile. Univer.s, de Chile, Inst. d. Geol., Publ. Nr. 14, Santiago, 1960.
10. Munoz Cristi, J. Geologia de Chile. In: Geografa Economica de Chile. Tomo I. Santiago
de Chile (1950).
11. Tazieff, H. Interpretation des glissements de terrain accompagnant le grand seisme du
Chile. Bull. de la Soc. Beige de Geologie, Tome LXIX fase. 3. 1961, S. 1-11.
12. Auer, V. Las capas volcanicas comp nuevo mdtodo de cronologia postglacial en Fuegopata-
gonia. Gaea VIII (1948) S. 311-344.

BIBLIOGRAPHY (not referred to in text.)

Bauer, P.-P. von. Das Erdbeben vom 22. Mai 1960 in der Provinz Llanquihue in Suechile.
Erdkunde XIV (1960) S. 288-297.
Bruggen, J. Fundamentos de la Geologia de Chile. Santiago, 1950.
Corporation de Fomento de la Production. Avaluo de los danos ocasionados por los sismos,
inundaciones y maremotos en la zona sur de Chile. Santiago, 1960.
Depto de Navegacion e Hidrografia de la Armada. El maremoto del 22 de mayo de 1960. Boletin
Inform. Valparaiso Alto XVI (1960) Nr. 56.
Dobrovolny, Lemke, Bowes, Thomas y Bravo. Relation entre la Geologia y el dano producido
por el terremoto del 22 de mayo de 1960 en Puerto Montt. Instituto de Investigaciones Geo-
logicas. Santiago de Chile (1960).
Doyel, Maraga and Falcon. Relaciones entre la Geologia de Valdivia y los danos causados por
los terremotos del 22 de mayo 1960. Inst. d. Inv. Geolog. Santiago 1960.
Guarda Geiwitz, F. Historia de Valdivia 1552-1952. Santiago de Chile, 1958.
Lotze, Fr. Notizen zur Aktuo-Geologie, Neues Jahrh, f. Geol. and Palaont. Monatshefte H.
11 (1960) S. 519-524.
Meyer-Rusca, W. Die chilenische Schweiz. Santiago 1950.
Robinson, A. H. W. The Pacific Tsunami of May 22nd., 1960. Geography Vol. XLVI (1961)
S. 18-24.
Veyl Onat, Carlos. Los fenomenos volcanicos y s仁icos de fines de mayo de 1960 en el Sur de
Chile. Univers-de Conception. Inst. Centr. de Quimica. Dpto. de Geol. y Min. (1960).
Weischet, W. Studien fiber den glazial bedingten Formenschatz der S・chilenischen Langssenke
im West-Ost-Profil beiderseits Osorno. Peterm. Mitt., Gotha 1958. S. 161-172.


WOLFGANG WEISCHET
INSTITUTO DE GEOGRAPHIA
UNIVERSIDAD AUSTRAL DE CHILE
Manuscript received on February 6, 1963.
Translated by Roland von Huene, U. S. Naval Ordnance Test Station, China Lake, Calif.