Field studies of granites and metamorphic rocks: Central ...
Stump, E. 1976. On the Late Precambrian-Early Paleozoic metavolcanic and metasedimentary rocks of the Queen Maud Mountains, Antarctica, and a comparison with rocks of similar age from southern Africa. (Ph.D. thesis, Columbus: Ohio State University.)
10 8
6 2
10
•
CTM Shockleton Coast
£ (TM Miller Range
NVL fidmiralty Intrusives J NVL Granite harbour Intrusives
0
Depleted mantle evolution band
8
O)
6
EN dit) 0
l+cmscoaiil
-2
.lflldMRGFllI
+
Ev olution trend Model age for continental 'for average crust in the Miller Range continental crustal Nd
-8 10 -12
: 0 100 200 300 400 500
Age (Ga)
Figure 2. This figure shows clearly the different isotopic signature of the granites on either side of the central Transantarctic Mountains (Miller Range [CTM MR] to the west and Shackleton Coast [CTM SC] to the east). The peraluminous chemistry of the granites in the Miller Range indicates that they are pure crustal melts and so their isotopic composition reflects the isotopic composition of the Precambrian crust in this area. The neodymium model ages (based on a depleted mantle) for these samples give an estimate of about 2.0 billion years for the average age of this Precambrian crust. The Shackleton Coast samples have considerably higher initial isotopic signature as well as model ages of about 1.5 billion years. These characteristics may be explained by mixing between mantle magmas and Precambrian crust similar to that which produced the Miller Range granites. However, we cannot rule out the possibility that the origin of the Shackleton Coast granites may involve a Precambrian component which is completely different from that represented by the granites in the Miller Range.
Field studies of granites and metamorphic rocks: Central Transantarctic Mountains S.C. BORG, J.W. GOODGE, and D.J. DEPAOLO Department of Earth and Space Sciences University of California Los Angeles, California 90024
J.M. MATTINSON Department of Geological Sciences University of California Santa Barbara, California 93106
This article summarizes field studies of the late Precambrian to early Paleozoic basement of the central Transantarctic Mountains (see Borg and DePaolo, Antarctic Journal, this issue). During the 1985-1986 season, we used helicopter support from the Beardmore Camp to conduct reconnaissance mapping and 1986 REVIEW
ESr(t) Figure 3. This figure allows a comparison of both samarium and neodymium compositions of the granites of the central Transantarctic Mountains with granites from northern Victoria Land (NvL). Most of the samples of Granite Harbor Intrusives from NVL. are peraluminous rocks, like those from the Miller Range, and probably represent pure crustal melts. As such, their isotopic composition reflects the isotopic composition of the lower crust in that region. From this data it appears that the Precambrian crust in the Miller Range is older than the Precambrian crust which melted to produce the Granite Harbor Intrusives in NVL. Also, the Miller Range granites involve crustal rocks which are much more rubidium-rich and have evolved considerably more radiogenic strontium. The Shackleton Coast granites are isotopically similar to the Admiralty Intrusives in NVL but, until we have filled out our coverage of isotopic ratios in the Transantarctic Mountains we cannot specify the precise nature of the Precambrian crust involved in producing them.
sampling in the segment of the range between the Nimrod and Good glaciers (figure 1). In addition, several areas were mapped in detail, including the Campbell Hills and Cape Lyttelton area, the Mount Hope and Cape Allen area, and a portion of the western side of the Miller Range. Our detailed geologic mapping represents a substantial improvement over published maps. A comparison of our maps with published geologic maps indicates clearly that much work is necessary to produce an accurate map of the basement rocks of the region. In the Campbell Hills and Cape Lyttleton area, over 50 percent of the area was incorrectly depicted in the American Geographical Society Map Folio series (Grindley and Laird 1969). Similarly, in the Mount Hope and Cape Allen area approximately 20 percent is incorrectly represented on the Mount Elizabeth and Mount Kathleen Geologic Quadrangle Map (Lindsay, Gunner, and Barrett 1973). We recognize that earlier work relied heavily on interpretation of aerial photographs and that ground information was often not available. However, because these maps are the basis for the geologic framework in which current work is founded, it is important to remind ourselves the extent to which this work represents inferences or extrapolated information. Granites. Batholithic rocks assigned to the lower Paleozoic Granite Harbor Intrusives in this region include a variety of lithologic types ranging from diorite to tourmaline-bearing, 243
[1 Regions of exposure of Poleozoic granites or undifferentiated granites and crystalline basement
Norftwn Victoria Land
Outcrop area of the Ross Sea Sector Th.s+ ROSS SEA
ck:i 'C
LF ROSS ICE SHELF
Byrd Gkxie
MARIE BYRD LAND
WEST ANTARCTIC //) ICE SHEET
EAST ANTARCTIC -- -- ICE SHEET Ni
Glacier
Scott Glacier 0 K 2mies
SOUTPOLE
./ Thiel ,Mountain
Figure 1. Location map of the study area in the Transantarctic Mountains. Boxed area is enlarged in figure 1 of Borg and DePaolo (Antarctic Journal, this issue).
mica granite (sensu stricto). These rocks generally occur as scattered single plutons and clusters of several plutons. Our work supports previous suggestions that these are epizonal plutons because only narrow contact aureols were found. Furthermore, broad roof zones of plutons are exposed in the Miller Range and at Cape Lyttleton. The batholithic system shows lithologic variation from diorite and granodiorite along the Shackleton Coast to felsic, 2-mica granites in the Miller Range. The name Hope Granite had been applied to all these post-tectonic granites (e.g., Gunner 1976); however, we feel this term is inappropriate because it includes lithologically diverse and physically distinct plutons which are clearly not related to a single parent magma. Our sampling program yielded about 200 samples of granite and 100 samples of metamorphic country rock, totaling about 5.5 tons of rock. Initial cataloging, organizing and processing is in progress. See Borg and DePaolo (Antarctic Journal, this issue) for initial analytical results. Metamorphic Rocks of the Miller Range. Structural and petrologic relations of metamorphic rocks were studied in the central and southern Miller Range. Much of the original description by Grindley et al. (1964) of Nimrod Group rocks in eastern parts of the range need not be revised, but we present new structural interpretations which diverge significantly from those of Grindley (1972). At the head of the Argosy Glacier, exposures of high-grade Miller Formation (Gunn and Walcott 1962) and metamorphic units of the Nimrod Group are separated by a complex zone of mylonitic rocks referred to by Grindley et al. (1964) as augen gneiss. This unit, a granodioritic, porphyroclastic mylonitic gneiss, contains a weak compositional foliation, which dips moderately southwest, and a well-formed extensional mineral lineation, which trends approximately northwest-southeast. 44
Mylonitic fabrics become stronger gradationally upward through an exposed zone approximately 500 meters thick toward rocks of the Miller Formation and, at highest structural levels, are cut by much thinner zones (0.1-2 meters) of ultramylonite. When viewed in a plane normal to mylonitic foliation and parallel to lineation, a number of small-scale structures such as asymmetric S-C fabrics, rotated porphyroclasts, and folds indicate a top-to-the-southeast sense of shear in a direction parallel to mylonitic lineation. Fabrics which are well-displayed throughout the mylonitic granodiorite are present in well-recrystallized metamorphic rocks of the Miller Formation only in the vicinity of the mylonitic granodiorite; thus, shear deformation must post-date granodiorite emplacement and metamorphism of Miller Formation rocks. From these relationships we conclude: (1) the Endurance thrust of Grindley (1972) is in fact a distributed shear zone which displaced high-grade rocks of the Miller Formation to the southeast over other (younger?) rocks originally designated as part of the Nimrod Group; (2) rocks of the Miller Formation may not be equivalent to other geologic units assigned to the Nimrod Group; (3) the Aurora Formation augen gneiss is better characterized as a mylonitic gneiss derived from granodiorite; and (4) the mylonite zone described here is offset by a number of highangle faults (figure 2), but is nowhere tightly folded as suggested of the Endurance thrust by Grindley (1972). The absence of shear deformation in nearby Granite Harbor Intrusives suggests that the mylonite pre-dates Ross Orogeny plutonism.
SW
NE
exposed
LiI F 71
Nimrod Group
S-
7
not exposed
Miller Formation gneiss Marble and calc-schist Mylonitic granodiorite gneiss
NOT TO SCALE
Figure 2. Sketch of geologic relations in the upper Argosy Glacier area, western Miller Range. See figure 1 of Borg and DePaolo (Antarctic Journal this issue) for location of this cross-section. The Endurance Thrust is faulted but not folded.
We would like to thank the support personnel both at the Beardmore Camp and at McMurdo who made the Beardmore Camp a successful venture and we would like to thank the Beardmore Camp helicopter detachment of VXE-6 for many excellent days of logistic support. This research was supported by National Science Foundation grant DPP 83-16807. References Grindley, G.W. 1972. Polyphase deformation of the Precambrian Nimrod Group, central Transantarctic Mountains. In Adie, R.J. (Ed.), Antarctic geology and geophysics. Oslo: Universitetsforlaget. ANTARCTIC JOURNAL
Grindley, G.W., and M.G. Laird 1969. Geology of the Shackleton Coast. American Geographical Society, Antarctic Map Folio Series, no. 12, plate 15. Grindley, G.W., and V.R. McGregor, and R.I. Walcott. 1964. Outline of the geology of the Nimrod-Beardmore-Axel Heiberg Glaciers region, Ross Dependency. In Adie, R.J. (Ed.). Antarctic geology. Amsterdam: North Holland. Gunn, G.M., and R.I. Walcott. 1962. The geology of the Mt. Markham
Sedimentology of the Pagoda Formation (Permian), Beardmore Glacier area J.M.G. MILLER and B.J. WAUGH Department of Geology Vanderbilt University Nashville, Tennessee 37235
The Permian Pagoda Formation in the Queen Elizabeth and Queen Alexandra Ranges and in the western Queen Maud Mountains records glacial, glaciofluvial, and glaciolacu strine conditions. Preliminary facies analysis of 15 Pagoda sections visited during November and December 1985 suggests that episodes of glacial advance and retreat can be recognized within the formation. Lithofacies in the Pagoda Formation include diamictite, sandstone, and shale (see also Lindsay 1968, 1970). The predominant facies is arenaceous diamictite with 10-15 percent clasts; this facies commonly contains deformed sandstone inclusions. Silty diamictite with less than 5 percent clasts occurs in some sections. Most diamictites are internally structureless; winnowed levels, boulder pavements, and striated and grooved surfaces occur locally (figure 1). Sandstones are either coarse-grained, pebbly, and trough and planar cross-bedded; or fine- to medium-grained, rippled, and parallel-laminated; or massive. The sandstones form channel-fills within diamictite, tabular bodies interbedded on the meter scale with diamictite (figure 2), and beds and lenses, generally deformed, within shale sequences. Thick sequences of sandstone with abundant soft-sediment deformation occur locally. Shales (siltstone or sandy siltstone are occasionally parallel-laminated but most commonly u& structureless probably because of soft-sediment deformation. The shales locally contain scattered clasts and commonly ii: dude limestone concretions. The relative abundance of the lithofacies varies: some sections are dominated by diamictite, others by diamictite and sandstone, while shales are present only in certain areas. In general, shales are more common at the top of the formation showing a gradation into the overlying Mackellar Formation. The diamictite units are interpreted as lodgment, melt-out, redeposited, and waterlain tills. Some sandstone beds were likely deposited by mass flow, others by glaciofluvial and 1986 REVIEW
region, Ross Dependency, Antarctica. New Zealand Journal of Geology and Geophysics., 5, 407-426. Gunner, J. 1976. isotopic and geochemical studies of the pre-Devonian basement complex, Beardmore Glacier region, Antarctica. (Institute of Polar Studies, Report #41, Ohio State University, Columbus, Ohio.) Lindsay, J.F., J. Gunner, and P.J. Barrett. 1973. Reconnaissance geologic map of the Mount Elizabeth and Mount Kathleen Quadrangles, Transantarctic Mountains, Antarctica. (U.S.G.S. Antarctic Geologic Quadrangle Map A-2.)
glaciolacustrine processes. The shales are most likely glaciolacustrine because of an absence of any marine characteristics. Soft-sediment deformation within shale units probably represents downslope slumping, whereas that associated with sandstone and diamictite, and locally with striated surfaces, likely formed by glaciotectonic processes. Lodgment till exists at the base of the formation at almost all localities visited. In places, ice-contact deposits dominate the whole Pagoda section, with some glaciofluvial interbeds. At other locations, glaciolacustrine conditions existed in middle and upper parts of the section. Rare directional indicators show transport toward the south and southeast (figure 1; see also Lindsay 1970). The relative abundance of meltwater deposits implies deposition under a temperate or humid subpolar climate (Eyles, Eyles, and Miall 1983). Episodes of glacial advance and retreat can be recognized through analysis of vertical facies sequences in the Pagoda Formation. In addition to facies sequence, critical features for inferring advance and retreat include: (1) grooved or striated surfaces (figure 1), (2) presence or absence of sandstone interbedded with diamictite and abundance and character of this sandstone, (3) diamictite character including evidence of shearing or reworking, (4) boulder pavements and concentrations, and (5) sharp sedimentary contracts. '7!
,
I
Figure 1. Grooved sandstone surface with striated boulders at one end. Ice moved to right, parallel to hammer handle. (Hammer is 45 centimeters long.)
45
10 8
6 2
10
•
CTM Shockleton Coast
£ (TM Miller Range
NVL fidmiralty Intrusives J NVL Granite harbour Intrusives
0
Depleted mantle evolution band
8
O)
6
EN dit) 0
l+cmscoaiil
-2
.lflldMRGFllI
+
Ev olution trend Model age for continental 'for average crust in the Miller Range continental crustal Nd
-8 10 -12
: 0 100 200 300 400 500
Age (Ga)
Figure 2. This figure shows clearly the different isotopic signature of the granites on either side of the central Transantarctic Mountains (Miller Range [CTM MR] to the west and Shackleton Coast [CTM SC] to the east). The peraluminous chemistry of the granites in the Miller Range indicates that they are pure crustal melts and so their isotopic composition reflects the isotopic composition of the Precambrian crust in this area. The neodymium model ages (based on a depleted mantle) for these samples give an estimate of about 2.0 billion years for the average age of this Precambrian crust. The Shackleton Coast samples have considerably higher initial isotopic signature as well as model ages of about 1.5 billion years. These characteristics may be explained by mixing between mantle magmas and Precambrian crust similar to that which produced the Miller Range granites. However, we cannot rule out the possibility that the origin of the Shackleton Coast granites may involve a Precambrian component which is completely different from that represented by the granites in the Miller Range.
Field studies of granites and metamorphic rocks: Central Transantarctic Mountains S.C. BORG, J.W. GOODGE, and D.J. DEPAOLO Department of Earth and Space Sciences University of California Los Angeles, California 90024
J.M. MATTINSON Department of Geological Sciences University of California Santa Barbara, California 93106
This article summarizes field studies of the late Precambrian to early Paleozoic basement of the central Transantarctic Mountains (see Borg and DePaolo, Antarctic Journal, this issue). During the 1985-1986 season, we used helicopter support from the Beardmore Camp to conduct reconnaissance mapping and 1986 REVIEW
ESr(t) Figure 3. This figure allows a comparison of both samarium and neodymium compositions of the granites of the central Transantarctic Mountains with granites from northern Victoria Land (NvL). Most of the samples of Granite Harbor Intrusives from NVL. are peraluminous rocks, like those from the Miller Range, and probably represent pure crustal melts. As such, their isotopic composition reflects the isotopic composition of the lower crust in that region. From this data it appears that the Precambrian crust in the Miller Range is older than the Precambrian crust which melted to produce the Granite Harbor Intrusives in NVL. Also, the Miller Range granites involve crustal rocks which are much more rubidium-rich and have evolved considerably more radiogenic strontium. The Shackleton Coast granites are isotopically similar to the Admiralty Intrusives in NVL but, until we have filled out our coverage of isotopic ratios in the Transantarctic Mountains we cannot specify the precise nature of the Precambrian crust involved in producing them.
sampling in the segment of the range between the Nimrod and Good glaciers (figure 1). In addition, several areas were mapped in detail, including the Campbell Hills and Cape Lyttelton area, the Mount Hope and Cape Allen area, and a portion of the western side of the Miller Range. Our detailed geologic mapping represents a substantial improvement over published maps. A comparison of our maps with published geologic maps indicates clearly that much work is necessary to produce an accurate map of the basement rocks of the region. In the Campbell Hills and Cape Lyttleton area, over 50 percent of the area was incorrectly depicted in the American Geographical Society Map Folio series (Grindley and Laird 1969). Similarly, in the Mount Hope and Cape Allen area approximately 20 percent is incorrectly represented on the Mount Elizabeth and Mount Kathleen Geologic Quadrangle Map (Lindsay, Gunner, and Barrett 1973). We recognize that earlier work relied heavily on interpretation of aerial photographs and that ground information was often not available. However, because these maps are the basis for the geologic framework in which current work is founded, it is important to remind ourselves the extent to which this work represents inferences or extrapolated information. Granites. Batholithic rocks assigned to the lower Paleozoic Granite Harbor Intrusives in this region include a variety of lithologic types ranging from diorite to tourmaline-bearing, 243
[1 Regions of exposure of Poleozoic granites or undifferentiated granites and crystalline basement
Norftwn Victoria Land
Outcrop area of the Ross Sea Sector Th.s+ ROSS SEA
ck:i 'C
LF ROSS ICE SHELF
Byrd Gkxie
MARIE BYRD LAND
WEST ANTARCTIC //) ICE SHEET
EAST ANTARCTIC -- -- ICE SHEET Ni
Glacier
Scott Glacier 0 K 2mies
SOUTPOLE
./ Thiel ,Mountain
Figure 1. Location map of the study area in the Transantarctic Mountains. Boxed area is enlarged in figure 1 of Borg and DePaolo (Antarctic Journal, this issue).
mica granite (sensu stricto). These rocks generally occur as scattered single plutons and clusters of several plutons. Our work supports previous suggestions that these are epizonal plutons because only narrow contact aureols were found. Furthermore, broad roof zones of plutons are exposed in the Miller Range and at Cape Lyttleton. The batholithic system shows lithologic variation from diorite and granodiorite along the Shackleton Coast to felsic, 2-mica granites in the Miller Range. The name Hope Granite had been applied to all these post-tectonic granites (e.g., Gunner 1976); however, we feel this term is inappropriate because it includes lithologically diverse and physically distinct plutons which are clearly not related to a single parent magma. Our sampling program yielded about 200 samples of granite and 100 samples of metamorphic country rock, totaling about 5.5 tons of rock. Initial cataloging, organizing and processing is in progress. See Borg and DePaolo (Antarctic Journal, this issue) for initial analytical results. Metamorphic Rocks of the Miller Range. Structural and petrologic relations of metamorphic rocks were studied in the central and southern Miller Range. Much of the original description by Grindley et al. (1964) of Nimrod Group rocks in eastern parts of the range need not be revised, but we present new structural interpretations which diverge significantly from those of Grindley (1972). At the head of the Argosy Glacier, exposures of high-grade Miller Formation (Gunn and Walcott 1962) and metamorphic units of the Nimrod Group are separated by a complex zone of mylonitic rocks referred to by Grindley et al. (1964) as augen gneiss. This unit, a granodioritic, porphyroclastic mylonitic gneiss, contains a weak compositional foliation, which dips moderately southwest, and a well-formed extensional mineral lineation, which trends approximately northwest-southeast. 44
Mylonitic fabrics become stronger gradationally upward through an exposed zone approximately 500 meters thick toward rocks of the Miller Formation and, at highest structural levels, are cut by much thinner zones (0.1-2 meters) of ultramylonite. When viewed in a plane normal to mylonitic foliation and parallel to lineation, a number of small-scale structures such as asymmetric S-C fabrics, rotated porphyroclasts, and folds indicate a top-to-the-southeast sense of shear in a direction parallel to mylonitic lineation. Fabrics which are well-displayed throughout the mylonitic granodiorite are present in well-recrystallized metamorphic rocks of the Miller Formation only in the vicinity of the mylonitic granodiorite; thus, shear deformation must post-date granodiorite emplacement and metamorphism of Miller Formation rocks. From these relationships we conclude: (1) the Endurance thrust of Grindley (1972) is in fact a distributed shear zone which displaced high-grade rocks of the Miller Formation to the southeast over other (younger?) rocks originally designated as part of the Nimrod Group; (2) rocks of the Miller Formation may not be equivalent to other geologic units assigned to the Nimrod Group; (3) the Aurora Formation augen gneiss is better characterized as a mylonitic gneiss derived from granodiorite; and (4) the mylonite zone described here is offset by a number of highangle faults (figure 2), but is nowhere tightly folded as suggested of the Endurance thrust by Grindley (1972). The absence of shear deformation in nearby Granite Harbor Intrusives suggests that the mylonite pre-dates Ross Orogeny plutonism.
SW
NE
exposed
LiI F 71
Nimrod Group
S-
7
not exposed
Miller Formation gneiss Marble and calc-schist Mylonitic granodiorite gneiss
NOT TO SCALE
Figure 2. Sketch of geologic relations in the upper Argosy Glacier area, western Miller Range. See figure 1 of Borg and DePaolo (Antarctic Journal this issue) for location of this cross-section. The Endurance Thrust is faulted but not folded.
We would like to thank the support personnel both at the Beardmore Camp and at McMurdo who made the Beardmore Camp a successful venture and we would like to thank the Beardmore Camp helicopter detachment of VXE-6 for many excellent days of logistic support. This research was supported by National Science Foundation grant DPP 83-16807. References Grindley, G.W. 1972. Polyphase deformation of the Precambrian Nimrod Group, central Transantarctic Mountains. In Adie, R.J. (Ed.), Antarctic geology and geophysics. Oslo: Universitetsforlaget. ANTARCTIC JOURNAL
Grindley, G.W., and M.G. Laird 1969. Geology of the Shackleton Coast. American Geographical Society, Antarctic Map Folio Series, no. 12, plate 15. Grindley, G.W., and V.R. McGregor, and R.I. Walcott. 1964. Outline of the geology of the Nimrod-Beardmore-Axel Heiberg Glaciers region, Ross Dependency. In Adie, R.J. (Ed.). Antarctic geology. Amsterdam: North Holland. Gunn, G.M., and R.I. Walcott. 1962. The geology of the Mt. Markham
Sedimentology of the Pagoda Formation (Permian), Beardmore Glacier area J.M.G. MILLER and B.J. WAUGH Department of Geology Vanderbilt University Nashville, Tennessee 37235
The Permian Pagoda Formation in the Queen Elizabeth and Queen Alexandra Ranges and in the western Queen Maud Mountains records glacial, glaciofluvial, and glaciolacu strine conditions. Preliminary facies analysis of 15 Pagoda sections visited during November and December 1985 suggests that episodes of glacial advance and retreat can be recognized within the formation. Lithofacies in the Pagoda Formation include diamictite, sandstone, and shale (see also Lindsay 1968, 1970). The predominant facies is arenaceous diamictite with 10-15 percent clasts; this facies commonly contains deformed sandstone inclusions. Silty diamictite with less than 5 percent clasts occurs in some sections. Most diamictites are internally structureless; winnowed levels, boulder pavements, and striated and grooved surfaces occur locally (figure 1). Sandstones are either coarse-grained, pebbly, and trough and planar cross-bedded; or fine- to medium-grained, rippled, and parallel-laminated; or massive. The sandstones form channel-fills within diamictite, tabular bodies interbedded on the meter scale with diamictite (figure 2), and beds and lenses, generally deformed, within shale sequences. Thick sequences of sandstone with abundant soft-sediment deformation occur locally. Shales (siltstone or sandy siltstone are occasionally parallel-laminated but most commonly u& structureless probably because of soft-sediment deformation. The shales locally contain scattered clasts and commonly ii: dude limestone concretions. The relative abundance of the lithofacies varies: some sections are dominated by diamictite, others by diamictite and sandstone, while shales are present only in certain areas. In general, shales are more common at the top of the formation showing a gradation into the overlying Mackellar Formation. The diamictite units are interpreted as lodgment, melt-out, redeposited, and waterlain tills. Some sandstone beds were likely deposited by mass flow, others by glaciofluvial and 1986 REVIEW
region, Ross Dependency, Antarctica. New Zealand Journal of Geology and Geophysics., 5, 407-426. Gunner, J. 1976. isotopic and geochemical studies of the pre-Devonian basement complex, Beardmore Glacier region, Antarctica. (Institute of Polar Studies, Report #41, Ohio State University, Columbus, Ohio.) Lindsay, J.F., J. Gunner, and P.J. Barrett. 1973. Reconnaissance geologic map of the Mount Elizabeth and Mount Kathleen Quadrangles, Transantarctic Mountains, Antarctica. (U.S.G.S. Antarctic Geologic Quadrangle Map A-2.)
glaciolacustrine processes. The shales are most likely glaciolacustrine because of an absence of any marine characteristics. Soft-sediment deformation within shale units probably represents downslope slumping, whereas that associated with sandstone and diamictite, and locally with striated surfaces, likely formed by glaciotectonic processes. Lodgment till exists at the base of the formation at almost all localities visited. In places, ice-contact deposits dominate the whole Pagoda section, with some glaciofluvial interbeds. At other locations, glaciolacustrine conditions existed in middle and upper parts of the section. Rare directional indicators show transport toward the south and southeast (figure 1; see also Lindsay 1970). The relative abundance of meltwater deposits implies deposition under a temperate or humid subpolar climate (Eyles, Eyles, and Miall 1983). Episodes of glacial advance and retreat can be recognized through analysis of vertical facies sequences in the Pagoda Formation. In addition to facies sequence, critical features for inferring advance and retreat include: (1) grooved or striated surfaces (figure 1), (2) presence or absence of sandstone interbedded with diamictite and abundance and character of this sandstone, (3) diamictite character including evidence of shearing or reworking, (4) boulder pavements and concentrations, and (5) sharp sedimentary contracts. '7!
,
I
Figure 1. Grooved sandstone surface with striated boulders at one end. Ice moved to right, parallel to hammer handle. (Hammer is 45 centimeters long.)
45
