20.On the Volcanism of the Huzi Volcanic Zone, with Special Reference to the Geology and Petrology of Idu and the Southern Islands. | 20.富士火山帯の火山作用,特に伊豆及び南方請島の地質及び岩石に就いて

When volcanic activity began in the region studied is a matter of conjecture. There is no doubt, however, that volcanoes were already in existence in Eocene times, seeing that the Ogasawara (Bonin) Islands are composed almost exclusively of andesitic rocks, both explosive and effusive, with a few intercalated Eocene limestones. If these islands were located on the submarine ridge that underlies the insular volcanoes of the Huzi volcanic zone, it would be more than likely that the Ogasawara volcanics are products of the earliest recognized forerunner of volcanic activity in that zone, but their more or less outlying position seem to militate against this view. However, notwithstanding the lack of positive evidence indicating structural similarity, the writer presents the above view as a possible explanation, because the Ogasawara andesites show the same general characteristics, both microscopical and chemical, as the allied rocks from the Huzi volcanic zone. The first indisputable record of volcanic activity in the Huzi volcanic zone dates to early Miocene. This seems to be the case not only with the Idu region, in the study of which the writer is chiefly concerned, but also with the much larger region of the "fossa magna", of which Idu forms but a small part. Through the whole of the period that succeeded it, the same general sequence of volcanic activity has beeu maintained in most parts of the fossa magna with certain local differences. It is therefore highly probable that, although local volcanic activity had already manifested itself in Eocene times in the Ogasawara Islands, more active and more regional activity had started contemporaneously with the tectonic disturbances, the result of all which is the fossa magna, which severed the geologic continuation between southwestern and northeastern Japan. As to whether or not the Ogasawara Islands were subjected to the same disturbances, the writer is unable to say. The lower Miocene volcanism in Idu, which occurred beneath the sea of that time, is represented by a thick series of highly altered volcanics (lavas and pyroclastics), referred to as the Yugasima beds (or the propylite series). This series consists for the most part of basaltic and andesitic rocks, although dacites and liparites are also represented, especially in the upper horizons of the series. The succession of these rocks implies considerable time, during which certain structural deformations might possibly have recurred, resulting in several periods of volcanic accumulation. But its actual duration is a question we have no means of answering at present, the only precise datum being the faunal evidence (Lepidocyclina limestone from Simo-Siroiwra and Miogypsina limestone from Nasimoto), which assigns certain dacitic volcanics lying some distance above the base of the series to lower Miocene. The propylite series is invaded by numerous minor intrusives (dikes, sheets, and necks) of various rock-types―diabase, porphyrite, basalt, andesite, dacite, and liparite (quartz-porphyry). Although these intrusives are evidently later than the propylite series, not all of them were formed contemporaneously. The basic intrusives (including diabase, porphyrite, basalt, and andesite) are mostly older, all being altered in the same way as the propylite series in which they are found, whereas the more acid intrusives are mostly somewhat younger, being probably correlated with the acid volcanics that overlie unconformably the propylite series. The lower Miocene volcanism was followed, probably during upper Miocene, by submarine eruptions of liparites and dacites, with subordinate andesites, all of'which are distributed mostly in south Idu and rather locally in north Idu. As just mentioned, some of the minor intrusives in the propylite series, especially those of acid rocks, might be the feeders of the upper Miocene volcanics. In contrast with the propylite series, these upper Miocene volcanics are either quite fresh or are but little altered. The foregoing phase of volcanism was succeeded, most probably during lower Pliocene, by a subordinate phase of activity characterized by minor intrusions (dikes) and extrusions of andesites. Regarding volcanism during this period, however, we are largely in the dark as only a few rocks in Idu have been identified with that period. In the uppermost Pliocene, eruptions of basalts, dacites, and andesites, occurred only locally in Idu, and these were again exclusively submarine. At the close of the Tertiary or the beginning of the Pleistocene, Idu peninsula was subjected like the other regions in the fossa magma to crustal movement, generally in the sense of uplift, and it was not until some time after this movement, which raised so much of the Tertiary sea-floor into land, that Pleistocene volcanism became vigorous. This volcanism was concentrated mainly in the northern part of the peninsula, while in the southern part, Tertiary volcanism was succeeded by that relative quiescence which marked the Pleistocene aud Recent epochs. The first phase of lower Pleistocene volcanism is thus represented by eruptions of basalts, andesites, and dacites, all occurring in the forms of tuff and tuff-breccia that now occupy the bulk of the lacustrine deposits (Ono beds, Simo-Hata beds, and Simo-Tanna beds) accumulated during that period on the erosion reliefs of the Tertiary formations. It was followed at the end of lower Pleistocene and in the upper Pleistocene, when the lower Pleistocene deposits had been more or less deformed, by eruptions of central volcanoes. The volcanoes Tensi, Usami, Taga, Yugawara, Asitaka, Amagi, Hakone, and others in the northwestern part of the Idu peninsula yet unstudied, as also the dacitic masses in the vicinity of Atami, are ^ all products of this phase of activity, with however certain differences in the times when the activities began and in the period for which they lasted. The basaltic- Omuro-yama group represents the last volcanism in the peninsula, which probably occurred some time later when the older volcanoes Tensi, Usami, Taga, etc., were subjected to more or less dislocation and dissection ; while the numerous hot-springs there represent a stage of post-volcanic thermal activity. So far as its visible part is concerned, Huzi may date from a much later age, probably Recent. That one of the earliest lava-flows from the volcano is younger than the older somma of Hakone has been definitely proved by field evidence. All the insular volcanoes of the Seven Idu Islands and the Southern Islands are undoubtedly products of Quaternary volcanism, whatever may be the history of their foundation. Of these, the dissected basaltic volcanoes Utone-sima, To-sima, Mikura-sima, Higasi-yama (Hatizyo-sima), Tori-sima, and Kita lo-sima probably date to Pleistocene. With the exception of Tori-sima111), we have no record of eruption of any of them. The basaltic volcanoes 0-sima, Miyake-sima, Nisi-yama (Hatizyo-sima), and Aoga-sima have been active in Recent times, all displaying lava out-flows in historic times. The liparitic volcanoes Nii-sima and Kodu-sima, according to records, also have had one eruption each. With the exception of the rarer alkaline rocks (potash-liparite and trachyandesite), the rocks that erupted in the course of the volcanism just outlined above have all several characteristics in common, mineralogical as well as chemical, with however certain differences due to local peculiarities. Whether chemically or mineralogically, there seems to be no systematic difference between the Tertiary and Quaternary series, excepting that the former usually has a higher water-content than the latter. A petropraphic similarity in the Ogasawara rocks to allied rocks from Idu is quite apparent. Of the various rock-types represented, the basalts and basaltic andesites are most predominant besides being regional in distribution, while the others are rather local. These regional rock-types are usually characterized by phenocrysts of calcic plagioclase (anorthite to calcic labradorite) and pyroxenes (augite or hypersthene or both), with or without olivine, in a groundmass containing less calcic plagioclase (bytownite to sodic labradorite), pigeonite, magnetite, and one or more of silica minerals (cristobalite, tridymite, and quartz). It is notable that most of the young lavas from Miyake-sima, Hatizyo-sima, Aoga-sima, etc., are poor in phenocrysts or almost aphyric, while the old lavas from these volcanoes are sometimes strongly porphyritic with large phenocrysts of anorthite. Like the latter, most of the lavas from the Pleistocene basaltic volcanoes (Kita lo-sima, Tori-sima, Mikura-sima, etc.), as also the basalts in the Tertiary series, are strongly porphyritic, sometimes with large phenocrysts of anorthite, the less porphyritic or almost-aphyric rocks being rarely met with. The variation diagram of the aphyric and aphanitic rocks (magma variation diagram) from both the Tertiary series and the Quaternary volcanoes shows smooth curves for all oxides, showing the probable variations in the magmatic liquids from which most of the volcanic rocks of the present region seem to have been derived. The variation diagram of type-averages of the Tertiary porphyritic rocks is identical with the same diagram of the Quaternary porphyritic rocks. The two are again similar to the variation diagram of the well-differentiated rocks of one volcano, namely, Amagi, with the result that either of these two diagrams will represent the chemical variation for rocks of the entire region. The variation on the whole follows that of type-averages of the Japanese volcanic rocks, thus showing the common characteristics that distinguish them from corresponding type-averages of the world's igneous rocks: namely, in containing higher lime, lower alkalies, and larger ratios of iron-oxides to magnesia. But, as compared. with type-averages of the Japanese volcanic rocks, the corresponding type-averages of the present region are usually a little higher in soda but lower in potash. So also do the latter as compared with the rocks from the two other volcanic zones in Japan, namely, the medial volcanic zone of NE. Japan and the Ryukyu volcanic zone. The rocks of the region studied, except a few alkaline rocks, may be treated as a connected suite, all belonging to what Peacock calls the "calcic series", with high alkali-lime index, 62・5 for the Tertiary series and 65・7 for the Quaternary series. If the entire suite has originated from a common magma, we would have to admit that acid rocks such as liparites were derived by processes of differentiation from a basaltic magma. During the whole period since the Tertiary, however, volcanism in the present region has been alternately basaltic (or andesitic) and liparitic (or dacitic). This recurrence of similar rock-types points to one of two explanations : (1) either a body of acid magna that had separated out during early Tertiary or before it had remained throughout succeeding ages as an available source of the acid rocks that erupted during several distinct epochs, or (2) that differentiation to an advanced stage had proceeded repeatedly along definite lines at wide intervals of time. Although, of these two alternatives, the former is the more simple, the latter process also must have operated at certain local centres in the region, seeing that well-differentiated rocks intimately related, structurally as well as petrologically, occur in certain central volcanoes such as Amagi. Although the rocks of the so-called Huzi volcanic zone may be treated on the whole as a connected magmatic suite, we find certain differences among them due to local peculiarities, mineralogical as well as chemical. Thus the Huzi rocks differ from the allied rocks of Idu and the Southern Islands in that the former are distinctly higher in alkalies, especially potash, and lower in lime than the latter. So also do the rocks from the volcanoes Kayaga-take and liduna-yama compared with the allied rocks of Idu and the Southern Islands. Although these two volcanoes lie farther north of Huzi, they belong in a wide sense to the Huzi volcanic zone. From these facts in view, the writer believes that the Huzi volcanic zone may be divided petrologically into two subzones, namely, the northern inland zone referred to as the Huzi Volcanic Zone proper, and the southern oceanic zone referred to as the O-sima Volcanic Zone. The Huzi volcanic zone proper includes the inland volcanoes Huzi, Kayaga-take, liduna-yama Kurohime-yama, and many others (Myoko-zan, Yatuga-take volcano group, etc.) yet unstudied, all of which are linearly arranged in a NNW.-SSE. direction on the fossa magna, while the 0-sima volcanic zone includes, in addition to the volcanoes of the Idu peninsula, the insular and submarine volcanoes of the Pacific―the Seven Idu Islands and the Southern Islands. To generalize, although the rocks of the so-called Huzi volcanic zone, with the exception of a feiv alkaline rocks, belong on the whole to a calcic series, the rocks from its northern inland part―the Huzi volcanic zone proper―are higher in alkalies and loiver in lime than those from the southern oceanic part―the O-sima volcanic zone. Whether or not the chemical differences of these two sub-zones are in any way connected with some differences in the geotectonic structure and in the nature of rocks of their foundation is a question we have no means of answering at present. The O-sima volcanic zone just proposed has two kinds of alkaline rocks, namely, potash-liparite (Manzo-yama and Kodu-sima) and trachyandesite (Io-sima). As to the genesis of the potash-liparite, a possible explanation was offered in an earlier paper112). As to the genesis of the trachyandesite, little can be said with certainty at present except that its local distribution in the Volcano Islands group and the occcurrence in it of porphyritic olivine and augite, both of which are similar in microscopic characters to the respective minerals in the basalts from the same group, make it possible that it is genetically related to these basalts. In this connection the distribution of the sites of volcanic vents in the Volcano Islands group calls for notice. Although the general trend of this group is from NNW. to SSE.―that is, the trend of the so-called Huzi volcanic zone―, lo-sima, besides being elongated in outline from NE. to SW., has numerous solfataras which are aligned on fissures running in the same direction. The trachyandesitic submarine volcano, "Sin-to Volcano", also lies on a line running NE.-SW. through the basaltic volcano, Minami lo-sima. Similar relation holds between the basaltic and liparitic volcanoes of the Seven Idu Islands11'0. Therefore, if the sites of volcanic vents were fixed by dislocation or lines of weakness in the earth's crust, these two trends―the NNW.-SSE. trend of the basaltic (or andesitic) volcanoes and the NE.-SW. trend of the liparitic (or dacitic) and trachyandesitic volcanoes―may be a surface manifestation of the underground structure of the region, the former representing a principal fracture (Daly's "abyssal fissure")114) connected with deep-seated reservoirs of a regional basaltic magma, and the latter a subsidiary fracture connected with shallow-seated reservoirs of local differentiates, either liparitic or trachyandesitic, from the basaltic magma. In conclusion, a few remarks should be added here regarding the alteration through secondary changes of the Tertiary volcanics in Idu, although on this point the writer has not yet been able to gather much information. As already noted, the only rocks that are altered to any important extent are these of the propylite series and the minor intrusives in that series. The common secondary minerals in them are chlorite, calcite, sericite, epidote, limonite, haematite, pyrite, zeolite, albite, and quartz, with amorphous silica, two or more of these minerals always occurring together. The alteration, however, has not advanced to the same degree and in the same way throughout the series. In some parts of the series, most of the rocks are almost unchanged, although, as a rule, incipient alteration of the constituent minerals is found only to a very slight extent, while in many others the alteration is considerable, resulting in virtual disappearance of the original structure of the rock, whether viewed microscopically or in the field. We often find that complete recrystallization has occurred, showing a mosaic of quartz and feldspar, with occasional flakes of sericite and chlorite. Since the alteration of the propylite series has advanced farther not only at and near the contacts with minor intrusives in the series, but also in the intrusives themselves than at distances away from the contacts, it could only have been caused by local and chemical effects, that, is to say, by chemical reactions between the original rocks of the series and some hydrothermal solutions115'that possibly generated in a post-magmatic stage subsequent to the minor intrusions rather than by regional and dynamic agencies. The metamorphic changes justifying the name, propylite, were probably caused by a sodiferous hydrothermal solution, possibly related to the basic intrusives (diabase, porphyrite, basalt, and andesite). The product of this intense mineralization which is so economically important and which has been worked in several parts of the propylite series in the region, was generated probably by a hydrothermal solution rich in silica generally related to the acid intrusives (dacite and liparite or quartz-porphyry). K. Sugi divided the metamorphism of the Misaka series of the Nakagawa district adjoining the northern part of Hakone, which is intruded by a large quartz-diorite mass, and which is a lower Miocene series correlated with the propylite series of Idu, into two stages: dynamic metamorphism caused by a mountain-building movement generated within a geosynclinal area before the intrusion of the quartz-diorite and contact metamorphism at the time of that intrusion116). The writer is of the opinion that the dynamic metamorphism, if it did occur, must have been due merely to local stresses that generated outside the boundary of the quartz-diorite mass at the beginning of the intrusion of that mass, seeing that the metamorphism is quite local in the Misaka series both at and near the contacts with the intrusive, and that such metamorphism is never met with in the propylite series of Idu, which belongs to the same geotectonic unit as the Misaka series.