Solifluxion and Periglacially Modified Landforms at Wellington, New Zealand

Transactions and Proceedings of the Royal Society of New Zealand, Volume 82, 1954-55, Page 1001

Solifluxion and Periglacially Modified Landforms at Wellington, New Zealand C A. Cotton and M T. Te Punga Victoria University College, Wellington [Read before 8th N.Z. Science Congress, May 18. 1954; received by the Editor, May 24 1954.] Abstract Part I—Solifluxion on the Periglacial Landscape Soliflexion deposits, commonly consisting of frost-riven angular fragment of greywacke set in a matrix of silty sand, are widely spread in south-west Wellington. They were formed in a periglacial environment during later Southern Hemisphere glaciations. the latest of which h may be assumed as synchionous with the Mankato glaciation of North America Solifluxion was therefore probably active in south-west Wellington until about 8 000 years ago An extensive mantle of solifluxion debits has served as the parent material of the present soil. Colonization of the area by forest vegetation cannot have begun until solifluxion ceased. Part II—Periglacarlly Modified Landforms Minor relief on remnants of a more or less level upland Surface surviving on summits and ridge crests near Wellington has been modified il not wholly incised below the surface of an initial peneplain by periglacial processes involving solifluxion probably alter nating with episodes of vertical stream corrasion. At the margins of plateau remnants the characteristic features are broad and shallow “dells” of gentle declivity hanging above the heads of steep-grade, deeply-cut ravines that dissect the flanks of the uplands. Though somewhat resembling nivation cliques, these forms quite probably did not originate as such. They may have been scooped out by streams of solifluxion debris is Such causation has been described in Central Europe, where some valley features very likely developed by such a process have been also, but probably erroneously, attributed to corrasion effected by streams of regolith moving; spontaneously—a gravity-controlled process clamed by W. Penck as aclimitic and universal. Part I Solifluxion on the Periglacial Landscape (M. T. Te Punga) Contents Introduction. Previous work. Solifluxion deposits an example.  A section at Belmont. Bedrock.   Solifluxion layer S1.   Solifluxion layer S2. Topsoil and subsoil. Microscopic examination.    Bedrock.    Solifluxion layer S1.    Solifluxion layer S2.

Discussion of solifluxion in south-west Wellington. Nature of the solifluxion debris (or head) Prof of movement. Conditions of movement. Distribution. Age of solifluxion deposits. Genera geological considerations. Solifluxion debris as a stratigraphic marker. Modification of the landscape by solifluxion. History of the present soil, flora, and fauna. Introduction A Revised and reduced estimate of the time that has elapsed since the beginning of the last great retreat of the Pleistocene ice sheets has been made by Antevs (1953), who has attempted, though unsuccessfully, to reconcile his figures with those arrived at by the radiocarbon method. Antevs dates it about 19,000 B.P. (before the present). According to radiocarbon dating, however, the last glacial advance reached its maximum only 11,000 years ago. If Nev Zealand was glaciated as recently as this, the present landscape of the southern portion of the North Island of New Zealand should retain traces of a periglacial history. Field work in south-west Wellington has shown that evidence of periglacial conditions is abundantly preserved, mainly in the form of extensive solifluxion deposits on slopes, accompanied by congelifraction phenomena on level ground. Less commonly periglacial wedge structures can be found. Solifluxion deposits, with which this essay is mainly concerned, are so common that they invite reflection on a hypothesis advanced by Mortensen (1953) to the effect that during the ice ages temperature inversion produced colder condition in certain low-lying periglacial regions than in high-altitude glaciated regions Whether or not Mortensen's hypothesis can be applied to New Zealand, it is now apparent that previous concepts of the developmental history of the present, topography, soil, flora, and fauna of south-west Wellington require revision n the light of new discoveries and new interpretations of “slope deposits.” Previous Work The reader is referred to a paper by Willett (1950), entitled “The New Zealand Pleistocene snow line, climatic conditions, and suggested biological effects”, for a discussion of climatic conditions during the Pleistocene glaciation of New Zealand, including specific reference to the glacial and periglacial history of the southern part of the North Island. After critical consideration of the probable position of the Pleistocene snow line in New Zealand, Willett (loc. cit, pp. 44–6; agrees with Adkin (1912) that the highest, “central part of the Tararua Range (lit 41° S.) suffered minor valley and corrie glaciation during Pleistocene times”. He estimates that an area of 57 square miles was there above the Pleistocene snow line, which stood at 3.940 feet above present sea-level. In a discussion of periglacial areas, Willett states: “The greywacke hills of … the Tararnas and Rimutakas would have been covered with alpine tussock on their lower parts and bare above. Frost action and solifluxion would have gone on apace on the ranges, resulting in a large supply of greywacke scree “In his map (loc. cit., p. 33. Fig. 6) extensive areas in south-west Wellington are shown, however, as “forest zone”.

Solifluxion Deposits: An Example A Section at Belmont (Fig. 1 and Plate 33, Figs. 1, 2, 3) Locality: Road-cutting, 2.6 miles north-westward along Hill Road from Belmont Railway Station; grid reference N.160/446373, about 1,050ft. above sen-level. (See also Fig. 2, “Belmont Plateau.”) The section exposes the following beds:— 12in. topsoil and subsoil; 3ft. lo 10ft. (plus) layer of dominantly finegrained solifluxion debris (referred to as “solifluxion two”, S2). Unconformity 2ft. to 3ft. layer of coarse, rubbly solifluxion debris (referred to as “solifluxion one”, S1). Unconformity 4ft. (plus) bedrock. Detailed description of the section (in ascending stratigraphic order): Bedrock The bedrock consists of even-textured, medium to coarse-grained greywacke of Permian-Triassic-Jurassic age. This rock now shows three prominent sets of joints, approximately normal to one another, which divide it into roughly cubical and rectangular blocks. The spacing of the joints is such that blocks with sides measuring from 3in to 9in. are the most common. The bedrock is so thoroughly weathered that a hand specimen can be crushed easily with the fingers. Each joint-bounded block is fawnish-yellow, patterned with more or less concentric, iron-enriched, rusty bands due to “Liesegang” effects. The rusty layers, numbering: about twenty to au inch in the outside shell of each block, vary in colour from orange to dark brown, contrasting vividly with the fawnish-yellow body of the block. The bedrock shows no decrease in intensity of weathering at the base of this exposure. (Exactly similar, thoroughly weathered, fawnish-yellow bedrock, with orange and dark brown “Liesegang “bands, has been observed in a nearby section, where its thickness is up to 10ft.; and, although there is a very slight decrease in the intensity of weathering at the base of this, a hand specimen can still be crushed in the fingers with little difficulty.) Solifluxion Layer S1. The solifluxion debris of S1 overlies the weathered bedrock with marked unconformity. The contact between bedrock and S1, exposed for about one chain in the cutting, is remarkably sharp (Plate 33, Figs. 2, 3). S1 is a coarse rubble deposit, consisting mainly of angular blocks, flakes, and chips of very hard greywacke. The angular blocks range from 3 to 8in. and the chips from ½ to 3in. Flakes, which are pieces with one, or rarely two, major curved faces resulting from frost-riving of spheroidally weathered boulders, are commonly from 1 to 2in. thick and 3 to 6in. across. This coarse rubble, com-praising about 80 percent of the layer, is unsorted and is set in a matrix of fawnish light-brown silty sand with some clay. Rare subangular, small fragments of the bedrock are found in the basal 12in. of rubble; and in a few places near the contact with bedrock the matrix of the basal 3in. of S1 contains a small amount of sand derived from the bedrock. The blocks, flakes, and larger chips forming the rubble are composed mainly of comparatively fresh mid-grey and bluish-grey greywacke. The mid-grey pieces have a light-grey, bleached, weathered shell about 3/16in. thick; and the bluish-grey pieces have a similar shell that is faintly stained with rust. The larger blocks, flakes, and chips are so hard that

a vigorous hammer-blow is necessary to break them. The weakly cemented matrix in which they are set is, however, very soft, and collapses so readily that it is difficult to collect a hand specimen. Some, probably most, of the matrix material was already highly weathered before transport of the soliftuxion debris, and some has been weathered since movement ceased. Except in the top few inches there is no rusty staining or mottling, the colour of the matrix being even and constant throughout. An orange-brown, iron-stained pan-like zone, about ⅛ to ¼in. thick is present at the top of S1. The upper 1/16in. of this is a deep ruddy-brown colour, but the colour becomes progressively lighter downwards, and at the rather poorly defined base the zone fades to a pale orange. The soft pan-like zone is only slighly more cemented than the underlying material of S1 and the overlying material of S2. It is evenly developed where the contact between S1 and S2 slopes at more than 20°, but is unevenly developed where the contact slopes more gently; locally it is replaced by an ill-defined rust-stained zone about 2 to 3in. luck at the top of S1. Rare, tiny rhizomorphs penetrate the upper 6in. of S1 but they are not present in the basal 30in. Solifiuxion Layer S2. The solifluxion layer S2 unconformably overlies both S1 and the bedrock. About 90 per cent of the layer is composed of clayey silt with a small amount of fine sand This material is mid-brown in colour, the colour being even and constant in the basal 9ft. where S2 is thickest. It is of deeper colour and finer texture than the matrix of S1. Where S2 is 10ft. (plus) in thickness, there is near the base a 12m. layer of coarse angular greywacke rubble similar to that of S1, but with less rubble and more matrix. In the basal 5ft. there are a few small, ill-defined pockets and lenses of angular greywacke chips (½ to 1 ½in.), which give this part of S2 a rudely stratified appearance. Very rare isolated, tiny, angular chips are present in the upper 5ft. The weakly weathered outer shells of the larger rubble fragments in the coarse basal layer are similar to those of S1. Ramifying ferruginous rhizomorphs are abundant throughout S2. At a depth of 9ft. rhizomorphs with a diameter of ⅛in. are not uncommon, and finely-branched smaller ones penetrate right to the base of the exposed section. In some cases the original rootlet has been replaced by a delicate thin-walled tube which is filled with soft, powdery material, both the tube and its filling being dark reddish-brown in colour. The rhizomorphs penetrate the rubble layer near the base of S2, their paths being confined to the fine-grained matrix. Where S2 rests on S1 it can be observed in some places that the rhizomorphs in the upper 6in. of S1 are extensions from those in S2. S2 is the parent-material of the present soil. Topsoil And Subsoil. The soil of the area in which the section is exposed if mapped on Sheet 8 of the Soil Map of New Zealand (4 miles to the inch, coloured map, published by the Department of Scientific and Industrial Research) as “immature, from sedimentary deposits, yellow-brown loam subgroup, Crown loam group”. Microscopic Examination. Samples of the bedrock and of solifluxion layers S1 and S2 were examined with a binocular microscope (X 100). Bedrock: Examination of a freshly exposed fracture face of a sample of the bedrock collected 6in. below the contact with S1 showed that except for extremely rare flakes of white mica the only recognizable mineral fragments are those of quartz. The feldspar and femic minerals are decomposed entirely. A sample

of the bedrock was soaked for 12 hours in water, then crushed thoroughly between the fingers, and the clay and fine silt decanted.off with water. The residue, amounting to about 60 per cent of the washed sample, comprised small (⅛ to ¼in.) lumps of uncrnshed weathered bedrock (about 75 per cent), and fine quartz fragments (about 25 per cent). Many of the small, roundish lumps are cemented with secondary iron compounds, and in a few angular lumps it can be seen that the ferruginous “Liesegang” bands are sometimes lightly cemented, although usually they are no harder than the material which they stain. Several types of quartz are present —viz., colourless transparent, yellowish transparent, smoky–brown transparent (these three have a glassy appearance), milky translucent, yellowish translucent, and smoky-brown translucent (these three rather dull). The translucent and transparent types are about equally represented. The transparent quartz not uncommonly shows conspicuous conchoidal fracture. About 90 per cent of the quartz fragments are angular and about 10 per cent slightly subangular. Very rare fragments show two or three crystal Faces, but no complete idiomorphs could be found. The quartz fragments range in size from about 1/64 to 1/16in., the average size being about 1/32in. The translucent fragments are in general larger than the transparent. One of the rare flakes of white mica which were observed was roughly circular in outline with a diameter of 1/16in. and a thickness of about 1/100in. The centre of the flake was quite colourless and transparent, but a marginal translucent zone, about 1/100in wide, was lightly stained with rust. The cleavage was still perfect. The transparent central portion could be powdered easily by applying light pressure with a dissecting needle. Solifluxion One (S1): A sample from the basal 3in. of S1 was soaked in water, stirred vigorously, and then the clay and fine silt fraction decanted off. In the residue roundish lumps of thoroughly weathered bedrock greywacke (⅛in. to ¾in.) can be identified. They are similar in every respect to the roundish lumps of bedrock that remained nncrushed in the washed bedrock sample described above. These lumps, which are rather uncommon, are mixed with abundant small angular chips (⅛ to ½in.) of fine-grained, hard greywacke, most of which have conspicuous rusty weathered rims, but grey unweathered centres. Some of the smaller angular chips are weathered right through, but they are harder and of finer texture than the bedrock. A few tiny (1/10in.) dark-brownish, rusty, ferruginous nodules that cannot be crushed in the fingers are present; some of these have apparently developed by centripetal enrichment of greywacke fragments with secondary iron compounds. Other minute nodules (1/16 to ⅛in.) resemble ironstone nodules from the late-Pleistocene buckshot gravels of western Wellington (Te Punga, 1954, Fig. 1, p. 3). In rare cases it can be seen that nodules with centripetally developed centres have aecretionary (centrifugal) outer layers. The sand-size fraction of the matrix of S1 consists almost entirely of quartz fragments. All the types of quartz present in the bedrock are represented, but there are more transparent than translucent fragments, and the average grain size is noticeably smaller, fragments from 1/64 to 1/32in. being most common. There are no mica flakes in the sand fraction, although a few highly weathered mica flakes can be seen in the small lumps derived from the weathered bedrock. Solifluxion Two (S2): S2 contains no bedrock lumps, and is a much finer-grained deposit than S1. Small (1/16 to 1/4in.) sharply angular fragments of

almost unweathered fine-grained greywacke are quite common in S2. The sand-size fracion consists almost entirely of quartz fragments which are similar to those in in S1. The chief difference between S2 and S1 are shown in Table I. Table I Solifluxion One (S1) Solifluxion tuo (S2) Older Younger Unconformable on bedrock Uneconformable on S1 and bedrock Coarse rubble diminant sandy silt dommant Contains small lamps of bedrock No lumps of bedrock Matrix hander than S2 Matrix softer than S1 Matrix coarser than S2 Matrix finer than S1 Large curved flakes of spheroindally weathered boulders not uncommon Large eurved flakes very rate Weathered rims of greywacke fragment thicker then in S2 Weathered rims thinner than in S1 Ferrugminous modules more aboundant and larger than in S2 Xodules less common and smaller than in S1 Matrix is fawnish light-brown in colour Matrix is mid-brown in colour Rhizormorphs much less common than in S2, and confined to the top portion Rhizooutmorphs abundant throught A certain amount of S1 is probably merinoorporated in S2 Parent material of present soil Discussion of Solifuixion in South-West the Wellington The term “solifluxion” was proposed by J. G. Andersson (1906) for “tin: slow flowing from higher to lower ground of masses of waste saturated with water. “Andersson considered that solifluxion was not restricted to areas of any one type of climate, but maintained that a cold climate provided optimum conditions for its development. By common consent the term is now restricted to the description of cold-climate phenomena. Slopes as low as 1° to 3° are sufficient to bring about minor solifluxion movement (Black, 1951, p. 281), but significant downshope movement requires a slope of about 5° or more (Washburn, 1947) In the Canadian Arctic it has been noted that most solifluxion movement occurs “when the material is very mobile as a result of spring melting of ground-ice in the seasonally thawed zone overlying perennially frozen ground.” Indeed Salomon (1929), supported by Bryan (1946. p. 625), has restricted the use of the term soliflnxion to motion over a base of perennially frozen ground Budel (1994, p. 492) has pointed out that the phenomenon of solifluxion, or flowing ground is not confined to, or characteristic of, regions in which the ground is perennially frozen to a great depth Very favourable conditions for such processes are afforded when, or where, freezing of the ground in winter is followed in spring and summer by a long thawing period, extending over several months. A requisite condition is that during such period the melt-water from snow and ice remains in the soil, being unable to escape downward because the subsoil remains frozen and thus impermeable, and not being dried out by evaporation because of a low slimmer temperature A superficial sheet of material remains fully saturated, and the resulting slush, or slurry, will flow down quite gentle slopes Such soil flow is, moreover, favoured especially by conditions too cold for forest growth, such as certainly prevailed in Central Europe during ice ages.

In south-west Wellington, though it is almost certain that frozen ground must have played an important rôle in the solifluxion environment, it is still a matter for speculation whether the ground was perennially frozen as in modern tundra regions. Nature of the Solifluxion Debris (or Head) A characteristic feature of the solifluxion debris (termed “head” by English and French geologists) is the presence of sharply angular blocks, fragments, and chips of greywacke set in an even-textured matrix of sandy clay (Plate 33, Fig. 2, and Plate 34, Fig. 4). From considerations of the geological setting it seems certain that all the coarse angular materials, and indeed probably all the fine material except the clay and some fine silt, are the products of frost-riving (congelifraction). It is well known that the mechanical disintegration of rock, including its comminution into fine grains, by repeated freezing and thawing of water-saturated materials and the growth of ice crystals in joints, cracks, cleavage planes, partings, and pores, is almost the sole destructive process in cold climates. Chemical weathering is at a minimum in such conditions. Merrill (1904, p. 278) notes that frost action makes rock “finely fissile and pulverulent”; and the extensive comminution of rock into monomineralic grains and mineral fragments by frost action has been emphasized by Matthes (1900), Hobbs (1911), Högbom (1914), Lewis (19319), and Taber (1943). The angularity of all large fragments of rock, and of many mineral fragments, in the solifluxion debris of south-west Wellington cannot be explained satisfactorily by any process other than frost-riving. Proof of Movement That, the solifluxion debris has moved to its present site is beyond dispute, for the rock fragments in the solifluxion layers are commonly of materials quite distinct from those of the bedrock at the site of emplacement. Striking examples proving movement can be seen in cuttings on the Wellington-Wairarapa road over the Rimutaka Range, where purplish, dark red soli-fluxion debris, derived from pillow-lavas, overlies mid-grey greywaeke bedrock In fresh railway cuttings near Upper Hutt. small angular white fragments set in a brick-red matrix form sohfluxion debris-which unconformably overlies yellowish brown sand. (The brick-red and white materials are both unusual weathering products of greywacke) Commonly, however, proof of movement is provided by certain characters of the greywacke fragments of the solifluxion layer which show that they cannot have been derived from the immediately underlying greywacke bedrock. In many cases the greywacke fragments are conspicuously fresh and very hard, whereas the underlying bedrock is highly weathered and soft; in other cases the greywacke fragments are lithologically different from the underlying greywacke—e.g, fine-grained, banded chips overlying coarse-grained, massive bedrock; or coarse-grained fragments overlying fine-grained banded bedrock. In certain areas it is obvious that large flakes and blocks (up to 5ft.), riven from the huge (up to 10ft.) spheroidally weathered boulders of autochthonous block fields and now embedded in sohfluxion debris some distance from their source, have been transported to the site of deposition (Plate 34, Figs. 2, 5) A typical example of this can be seen in the Belmont district (Fig 2) Tn the vicinity of Belmont Road Trig groups of spheroidally weathered blocks are prominent on the crest of a discontinuous north-trending

strike ridge formed by a distinctive, coarse-grained, massive greywacke in which the only joints present may be spaced as much as 10ft. apart (cf. Plate 37, Fig. 1). To the east and west of this hard belt the greywacke is thin-bedded, fine-grained, and closely jointed, the major joints never being more than a few inches apart. During a long early period of weathering (probably pre-Pleistocene) the spheroidal boulders were formed in the regolith developing over the hard, coarse, massive rock with widely spaced joints, while the closely jointed greywacke was thoroughly softened by chemical decay to a considerable depth. Prior to the last period in which solifluxion took place differential erosion had led to the development of a prominent ridge formed of the greywacke with widely spaced joints. Under periglacial conditions the spheroidally weathered boulders crowning the ridge were flaked and fractured by frost action and the resulting fragments were carried downslope by solifluxion. Large flakes and residual blocks either whole or split by frost action close to Belmont Road Trig now form a conspicuous part of a solifluxion deposit almost a mile eastward from their source on the ridge. This deposit unconformably overlies the typical closely jointed, fine-grained deeply weathered greywacke. Conditions of Movement The hypothesis of movement by solifluxion in a periglacial environment seems to provide the only acceptable explanation of the distinctive characters of the deposits under discussion. Annual thawing of a frozen regolith, probably overlying perennially frozen ground, may be assumed to have afforded conditions suitable for the operation of this process. As solifluxion in south-west Wellington has now ceased and must have done so several thousand years ago, the nature of the movement and conditions of transport can only be inferred from features of the arrested streams, or stranded deposits, and in this connection a few relevant observations are recorded here. Solifluxion has taken place most commonly on slopes with angles ranging from 5° to 30°, and rarely on slopes as gentle as 2° or as steep as about 35°. Slopes greater than about 35° presumably exceeded the angle of repose for dried out solifluxion debris. The thickness of the solifluxion layer varies considerably. In the vicinity of divides it invariably thickens progressively downslope for some distance away from the crest (Plate 34, Fig. 3), and has thus reduced the pro-solifluxion relief. Thicknesses of from 3 to 12ft. are the most common. Deposits less than 3ft. thick are found near the crest of divides, and layers thicker than 12ft. near the toe of slopes, the latter do not commonly exceed 20ft. The heads of small valleys of the pre-solifluxion landscape have in many cases been choked with solifluxion debris, the greatest thickness so far observed in such a situation being about 50ft. In most cases the distance that material has been transported by sol (fluxion is difficult to estimate. In certain parts of the Belmont area movement of about a mile can be demonstrated, but the extent of proved movement more often ranges from only a few yards to about a quarter of a mile. The bedrock surface beneath the solifluxion material is commonly conspicuously smooth and clearly defined, and in many sections the boundary between the bedrock and the overlying solifluxion layer is remarkably sharp (Plate 34, Figs. 1, 4). There can be little doubt that the bedrock has been shaved by abrasion at the base of the moving solifluxion mass. Such shaving of the bedrock is particularly conspicuous where the solifluxion material contains far-travelled fragments

of greywacke that are much harder than the deeply weathered, and therefore soft, bedrock that has been overridden. As the pre-solifluxion landscape of south-west Wellington was characterized by extensive areas of deeply weathered greywacke, conspicuous examples of shaving beneath solifluxion material are very common. Even where the bedrock is as hard as the fragments in the solifluxion material, shaving has undoubtedly taken place. It is difficult to estimate the amount (depth) of shaving that has occurred, but it seems possible that several Fig. 1.—Map showing distribution of periglacial solifluxion deposits (stippled) in south-west Wellington. The position of the Pleistocene snow line, as estimated by Willett (1950). is also shown (heavy broken line),

feet of soft weathered bedrock may have been removed in this manner, and in the case of hard bedrock perhaps only a few inches. The “shavings” were incorporated initially in the basal part of the solifluxion layer, but eventually became mixed throughout the moving mass. In a section close to Mungaroa Trig, near Kaitoke, a series of small folds with wave-lengths of 9 to 18in. and amplitudes of 3 to 6in. has been developed in sandy material which overlies bedrock and forms the base of a solifluxion layer (Plate 35, Fig. 2). The folds are developed upslope from a small knob of hard greywacke projecting about 6in. above a smooth-shaven bedrock surface which here slopes at an angle of about 7°. Downslope movement was impeded by the projecting knob of greywacke bedrock, producing a drag on the material immediately upslope, which led to folding. These small folds seem to be similar to some of those termed “plications” by Bryan (1946, p. 634). It may be noted in passing that the sandy material which shows the folds appears to consist largely of locally derived “shavings”. The folded sand layer is overlain by 3ft. of fine angular rubble set in a matrix of sandy silt. The basal few inches of this rubble layer repeat the folds of the underlying sand, but the folds fade out vertically within 12in. above the bedrock surface. The materials forming a solifluxion layer are in general very poorly sorted, apart from a very common tendency for coarser fragments to be concentrated in the lower and finer materials in the upper part of each layer, as though the larger blocks had sunk during flow. In a few cases, however, a very crude stratification has been produced by the arrangement of the coarser fragments in poorly defined bands of limited length. Presumably this has been brought about by some form of sliding on slip pianos within the moving solifluxion mass. Distribution The map herewith (Fig. 1) shows the distribution of periglacial solifluxion deposits in south-west Wellington. Typical exposures may be seen in road cuttings in Wellington City (Plate 34, Figs. 3, 4), on Belmont Road (Plate 33. Figs. 1, 2), on Paekakariki Hill Road, on the Akatarawa Road, on the main highway over the Rimutaka Range, on the highway that crosses the divide between the Wairariapa depression and the Eketahuna part of the Manawatu drainage basin, and on the Pahiatua Track over the northern end of the Tararua Range. In a few places solifluxion deposits descend to within a few feet of present sea-level, and at a number of localities they may be found within 50ft. of it. For example, at Point Howard, on the eastern shore of Wellington Harbour, a solifluxion deposit was cliffed by the sea prior to the 5ft. uplift which accompanied the 1855 earthquake, and at Porirua Harbour a layer mantling low spurs that terminate in marine cliffs is commonly less than 50ft. above the sea. There can be little doubt that prior to the vigorous late- and post-Flandrian cliffing of the coast of southwest Wellington solifluxion material was not uncommon down to present sea-level, and possibly in some cases was present below it. Büdel (1944, p. 491) states that in Central Europe during the Ice Age solifluxion was active down to sea-level. It is even more significant that coastal hanging valleys in Portugal, only 40 degrees from the equator, in the anomalously warm North Atlantic region, have been shown by Guilcher (1949a) to be thickly plastered on their floors and side slopes with solifluxional head consisting of large, completely angular blocks of locally derived rocks set in a

sandy-clay matrix. The valleys are of the age of, and descend to, a eustatic shoreline at 30–35 metres, which is proved by local paleontological evidence to be of mid-Pleistocene (Tyrrhenian) age, so that the fill in them must be attributed to periglacial processes in one or more of the later Pleistocene glacial ages. This seems to be the most southerly record in the Northern Hemisphere of solifluxion on a large scale near sea-level. Guilcher (1948) has also reported that similar coastal hanging valleys contain fill of the same kind in Brittany, and has described (1949 b, p. 716) coastal Devon and Cornwall as “plastered” with “flows; of clay with blocks” to depths of 50 and even 100 feet. Age of Solifluxion Deposits The solifluxion layer which occurs so widely and which so commonly forms, the parent material of the present soil of south-west Wellington was no doubt formed during the final phase of Southern Hemisphere glaciation, and as the evidence available suggests contemporaneity of glaciation in both hemispheres this phase is; tentatively correlated with the Mankato stadial, the climax of which is dated at about 11,000 years ago. Two solifluxion layers are present at some localities, having been found for example at the Belmont locality that has been described, in the Wellington City area (Plate 35, Fig. 1), on the divide between the Wairarapa depression and the Eketahuna part of the Manawatu drainage basin, and in an area north of Porirua Harbour. The older layer may be assigned tentatively to a pre-Mankato stadial of the Wisconsin (= Würm) glaciation It is possible that the two solifluxion layers of south-west Wellington represent the two main phases of the Wisconsin glaciation. To date the writer has been unable to find any traces of a fossil soil formed on the older solifluxion layer during an interstadial that followed its accumulation. It may well be, however, that such soils have been completely removed by later solifluxion. General Geological Considerations Solifluxion Debris as a Stratigraphic Marker The solifluxion debris of south-west Wellington serves as a useful stratigraphic marker for the purposes of relative geological chronology, and also provides useful information for an absolute chronology. Its usefulness as a climatic indicator is obvious. The solifluxion deposit at Point Howard (east side of Wellington Harbour) was cliffed by the sea prior to the 1855 earthquake, which raised the land about 5ft. and brought marine cliffing to an abrupt end. It appears that there were no marine waters in the present Wellington Harbour area at the tune of the Mankato-maximum (11,000 years ago), when sea-level was about 300ft. lower than now. The cliffing of the deposit must therefore be related to the final stages of the Flandrian transgression. The geological setting is such that the solifluxion can safely be considered as immediately antedating the Flandrian and/or as early Flandnan. Because glacio-ustatic change of sea-level was a world-wide phenomenon, it may be argued that the deposit at Point Howard is of Mankato age. This may be cited as evidence in support of the hypothesis of contemporaneity of Southern with Northern Hemisphere glaciation. In the valley of the Korokoro Stream near its source solifluxion debris overlies alluvial gravel capping a low terrace cut in greywacke bedrock. A clean section here exposes 3ft. of angular solifluxion debris, overlying 4ft. of well-rounded

terrace gravel, unconformably overlying 25ft. of bedrock greywacke in which the Korokoro Stream is now incised. (Grid reference: N.160/439343.) It thus appears that this headwater stream has deepened its narrow gorge by 25ft. in not. more than 11,000 years. Modification of the Landscape by Solifluxion Modification of the landscape of south-west Wellington by solifluxion is discussed by Cotton in Part II of this paper. History of the Present Soil, Flora, and Fauna The present soil of south-west Wellington is in many extensive areas developed on solifluxion deposits. This development must have begun as soon as solifluxion ceased and the material dried out. Assuming that solifluxion was very active at the Mankato-maximum, perhaps only 11,000 years ago, it is a not unreasonable hypothesis that it continued until about 8,000 years ago (i.e., 3,000 years after the Mankato-maximum and 3,000 years before the thermal maximum). It is a fairly reasonable inference therefore that all the present soils within the periglacial portion of south-west Wellington have developed within the last 8,000 years. The general absence of “buckshot gravel” in soils derived from the greywacke of southern Wellington is almost certainly due to the fact that these soils are too young to have developed horizons of ironstone nodules (Te Punga, 1954). It must be borne in mind that the upper part of the solifluxion layer at the time its movement ceased contained a large amount of finely comminuted regolithic material much of which had already been thoroughly weathered prior to the onset of Pleistocene glaciation. A certain amount of this fine solifluxion material had formed part of earlier Pleistocene and perhaps Pliocene soils. The postglacial colonization of the barren solifluxion area by vegetation may therefore have been very rapid, for an excellent seed-bed awaited the colonizing plants. At an early stage in plant colonization the rigours of the climate may have prevented forest growth, but at the time of the thermal maximum the entire area could have been seeded from trees, giving birth directly to a forest vegetation in a manner that short-circuited the time-consuming processes required for the colonization of barerock surfaces by the complex lichen-to-tree plant successions. It is apparent that the forest trees which provided the seed for colonization of the tundra-like solifluxion wastes must have been growing beyond the periglacial zone, probably to the north and east of it, and possibly also to the west and south below present sea-level. The faunal history is closely linked with that of the flora, for each stage in faunal colonization must have been intimately related to a stage in development of the vegetation. Present distribution patterns of animals can only be interpreted satisfactorily after consideration of the points discussed above.