START MICROFILMED 1985 UNIVERSITY OF CALIFORNIA - BERKELEY GENERAL LIBRARY BERKELEY, CA 94720 COOPERATIVE PRESERVATION MICROFILMING PROJECT THE RESEARCH LIBRARIES GROUP, INC. Funded by . THE NATIONAL ENDOWMENT FOR THE HUMANITIES THE ANDREW W. MELLON FOUNDATION Reproductions may not be made without permission. ~~ CU-B THE PRINTING MASTER FROM WHICH THIS REPRODUCTION WAS MADE IS HELD BY THE MAIN LIBRARY UNIVERSITY OF CALIFORNIA BERKELEY, CA 94720 FOR ADDITIONAL REPRODUCTION REQUEST MASTER NEGATIVE NUMBER 95-3837 AUTHOR: G-reen, William Lowthian. TITLE: The volcanic. problem... ~ PLACE: Honolu lw pare: LC 188%] voLume DULL gs- CALL 22255 MASTER NO. NEG. NO. 3837 DU623 Green, William Lowthian G83 The volcanic problem from the point of view of Hawaiian X volcanoes. [Honolulu, 1884) 7p. 23cm. ® Caption- title. Presentation copy, signed by the author. 104897 Purchase J, Howell books 6/62 - SHELF LIST FILMED AND PROCESSEDBY LIBRARY PHOTOGRAPHIC SERVICE UNIVERSITY OF CALIFORNIA BERKELEY, CA 94720 sono. 816 0]0/6]2 pate 1/1 8|5 REDUCTION RATIO g 2 DOCUMENT ‘SOURCE THE BANCROFT LIBRARY | es [lo =i te — lx I 32 ” = uy 2 fl i = Ls 2s fs pe MICROCOPY RESOLUTION TEST CHART NATIONAL BUREAU OF STANDARDS STANDARD REFERENCE MATERIAL 1010a (ANSI and ISO TEST CHART No. 2) 7 \ J CM HYDE Ponocinin, THE VoLcANIC PROBLEM FROM THE POINT OF VIEW OF HAWAIIAN VOLCANOES. >» BY .W. L. GREEN, Captain C. E. Dutton, in his valuable paper on Hawaiian Volcanoes, which ac companies the Annual Report of the Di- rector of the United States Geological Survey for the year ending June 30, 1883, has called attention to certain phases of volcanic action on Hawaii which deserve the careful attention of all vulcanologists. At page 194 he observes: “They,” (that is, the volcanoes of Mauna Loa and Kilau- ea) “bring up more lava and dissipate more energy in the course of a century than any others of which we have knowl- edge. Their action is also the least ob- streperous. That they should give off the greatest amount of energy in the quietest manner, is surely a most curious and strik- ing fact,-but it does not imply anything radical or anomalous. gory of facts extreme cases are always deeply interesting and sometimes highly instructive.” , line Captain Dutton considers the usual hy- potheses which are offered as the main moving causes of volcanic action, or the power which raises and maintains vast col- umns of molten rock above the surface of the earth—as inadmissible. He reviews rapidly the hypothesis of the percolation of water to the heated rocks, as well as that of the original mixture of the molten rock with water; also the oxydizing or chemical hypothesis, and the crushing, heating and melting theory of Mallet, and finds each one entirely unsatisfactory; while the expansion of hot solid bodies on relief of pressure, and the assumption of local increments of heat, he dismisses as hardly worth refuting. : Captain Dutton however, in studying Hayaiian volcanoes adverts to a principle But in any cate- ° of action which I have long looked upon as a main factor not only of the action in these craters but one which illustrates the general principle of volcanic action the world over. At page 196 he observes: “In the Cal- deras of Mokuaweoweo and Kilauea, the lava lakes represent the summits of col- umns of liquid lava extending to depths in the earth which are probably very great. When great eruptions occur on the flank of the mountains at lower levels, these lakes are drained, and when the erup tions cease the lavas come up in them again, to their former level. But in the history of the mountains the heights of these liquid columns have not been con- stant, but have steadily increased. In Ki- lauea this has, to our knowledge, increased rather more than 400 feet in the last sixty years. Taking all the circumstances into consideration, it seems as if the summits of the columns tn each case were approximate equilih- rium levels dependent almost solely upon their mean density at any given time.” (The italics are mine.) Captain Dutton further advocates the probable existence of two important facts in any attempt at explanation of volcanic action; namely, a universal liquid or highly plastic substratum, and the fact that cold solid lava is denser than hot molten lava. If this be the relative condition of the earth’s solid crust and the liquid substra- tum, may not this difference of density be the simple “missing factor” which causes all lavas to appear at or above the surface of the earth? Captain Dutton seems however, to con- sider (page 184) that the hypothesis of the universal underlying magma being of less 2 density than the crust which rests or floats upon it, is not one that is admissible. In this view I am aware that he is sustained by one of the most able physicists and mathematicians of the day, Sir William Thomson, who in an address at the Glas- gow meeting of the British Association in 1876 has given it as his opinion that under such circumstances the fractured crust would sink “like a rammed ironclad” Taking all the facts as they probably exist however, it does not seem necessary to conclude that the denser crust of the earth should sink far into the less dense upper layer of the magma upon which it may rest; for, First. Itis admitted that the density of the liquid underlying magma—if it ex- ists—increases with the depth, and may increase very rapidly at first, both because the density must increase with the pres- sure, and because the iron and other heavy components, in all probability increase in quantity as the depth increases. Second. The earth’s erust even when divided into segments by fissures, must be supposed to be resting upon the liquid substratum very much as the arch stones of a bridge are placed, and as the hot liquid contracts more thao the cooler solid crust, from loss of heat, the simple weight of the crust tends to close the fissures, and the farther the crust descends, the closer the fissures tend to become. Third. The liquid into which the crust is supposed to be sinking is entirely dif- ferent to that into which the ironclad sinks, for the moment the molten lava becomes exposed to radiation in the atmosphere or even to loss of heat by conduction in a narrow fissure, it tends to freeze solid, heal the crack, and prevent any further sink- ing. But as a matter of fact we know that frac- tures do exist in the earth’s crust and the lava flows out over them sometimes in enormous floods, while at other times it builds up cones of solid rock and rises in conduits in the center of them far above the mean surface of the earth. It may be only a question of the relation between the density of the lava and the crust, how far the liquid column has to build itself up, to place everything in a state of equi- librium, and to prevent any further sink- ing of the crust, even if the increasing den- sity below did not do so. ; This principle is partly illustrated ina pool covered by ice in a heavy frost and upon which is a number of skaters. The mean density of the ice including the weight of the skaters may be greater than that of the water, and when a crack occurs in it the water will often appear above the surface, and yet in a heavy frost there is little danger, as the water freezes rapidly and thickens the ice or heals the crack. But, we are told that if the appearance of the molten substratum above the sur- face of the crust be due to the weight of the crust, the lava in each of the different volcanoes should stand at the same level, and the instance of the liquid columns in the conduits of Kilauea and Mauna Loa is constantly brought forward by geolo- gists to prove that inasmuch as they con- stantly stand with the surface of the one some 8,000 to 10,000 feet above the level of the other, it is impossible that they can be connected with the same interior liquid mass. A little reflection, however, shows that this does not necessarily follow. It is well known that a large portion of the upper and interior masses of Hawaiian mountains have a density when cold 3 about 2.90, whilst the more basic or olivi itic outflows on the lower slopes have very commonly a density of 3.10. If we may suppose that the same proportionate dif- ference exists in the density of the molten columns of lava in Kilauea and Mauna Loa, and assume the thickness of the earth’s crust beneath the top of Mauna Loa to be 25 miles, this alone would cause near- ly the difference in height of the two col- umns which actually exists. But indepen- dently of the difference in composition, the heat of the more powerful column of Mauna Loa, and which is usually a closed column that is not exposed to radiation, may be expected to be much greater than that of the Kilauea column, which is con- stantly exposed to radiation and general cooling. It will therefore be more ex- panded or less dense. A very simple ex- periment will illustrate how easily the levels of connected liquids may stand at different heights above the surface of a floating crust. We have only to take a block of wood, say three inches square on each face, and bore two half inch holes through it in which insert glass tubes pro- jecting, say two inches above one face of the block. Weight the block so that it will just float flush with the water level, the glass tubes projecting upward. Now our oil into one tube and alcohol into the other until the water is displaced in the 104797 tubes. It will be seen that these liqui stand in the tubes at difterent El far above the level of the water although all are connected below. This principle is well known but it seems to have been overlooked in the application of it to the earth’s floating crust. The prob able causes, however, of lava standing at different heights, although all connected with one liquid mass, are various. Among the most important is the size, shape and regularity or irregularity of the connect- ing opening. An artesian bore into a stra- tum of water started from a little rise may allow of a liberal outflow, whilst a spring situated many feet below but proceeding from the same stratum might allow of only a small trickle of water in conse- quence of the crookedness or contraction of the passage by which it comes up. A good e practical example of connected li- quids not standing at the same level is seen in our artesian wells in the neighborhood of Honolulu, where the water will always flow out liberally at a height of about 40 feet above the sea level, while there is no \.- doubt whatever that the same stratum of water is connected with the ocean and runs off into it at some depth beneath the surface, but there is a column of salt wa- ter at a specific gravity of 1.03 against that of the fresh water at 1. Besides which there is the extra length of pas- sage to force itself through between the spot where the artesian bore strikes it and the point where it reaches the sea. In fine, connected liquids even at the same density do not under all circumstances stand at the same level, and when the density varies they may stand steadily at very different levels. What is of still more importance in this question to re- member, 1s, that when two connected col- umns of lava change their heights by dif ferences occurring in their density, not the slightest symptom of sympathy be- tween them 1s to be expected from such a change. Any one may make some interesting ex- periments illustrative of the action of molten lavas in tubes connected with a general molten mass, by constructing a closed metal box in the top of which are inserted glass tubes of different diameters say 2 inches to 3 inch, and then nearly filling the whole with molten stearine. Apply lamps underneath so as to heat the stearine equally, but keeping the temper- ature under the boiling point. It will X then be found that the liquid stands stead- ily highest in the largest tubes while in the smallest ones it will stand the lowest. By connecting a narrow tube with the side of the box, by an elbow slanting down- wards, the stearine will often freeze low down in the tube. It is evident that the wider the tubes are, the less the contained liquid is cooled by radiation and conduc- tion in proportion to its mass. Besides which the convection currents act freely in the wider tubes, while they are almost destroyed in the narrow ones. The re- sult is a considerable relative excess of temperature and consequent expansion in the wider tubes so that a longer column of liquid is balanced by a shorter column in the narrow tubes, while in the case where the convection currents are de- stroyed, the liquid cools so rapidly as to set solid low down in the tube. These convection currents, be it observed have no connection with steam or expand- ing gases, but are the simple results of expansion by heat. In this experiment we can see before us how a liquid mass of one temperature and density below, may stand steadily at different heights in col- umns connected with it. The application to columns of lava connected with one uni- versal substratum standing at different heights in those columns is obvious. Mol- ten stearine seems to behave in many re- spects as molten lava does when exposed to cooling. A variety of interesting anal- ogles may be exhibited by the apparatus just described. By reducing the heat, or increasing the cold, a crust will form on the surface of the stearine. Allowing the whole apparatus to cool a little, the molten stearine will contract in the tubes, some- times retreating from the crust and leav- ing a space between. A second crust will occasionally form below, illustrating what is often seen in the crater of Kilauea. On slightly heating up the apparatus again and breaking up the crust it will be seen to descend with the convection currents and re-melt, just as the lava crusts appear to do in the lakes of Kilauea. In fine, while the behavior of the molten and solid stearine exhibits what most other experiments with molten substances do that differences of temperature’ may pro. duce a reversal of the relative densities in the solid and molten states, there is no diffi- culty about the formation of a solid crust on a molten substratum, while it suggests the possibility of even a denser crust re- 4 maining supported by a less dense liquid i it, Th when there is an op- portunity offered for building up tubes or cones above the crust, in which a higher column of the less dense liquid acting on the mass below becomes the substitute for its smaller density and the crust becomes hydrostatically balanced. Volcanoes such as Kilauea and Mokuaweoweo with their cones and conduits containing liquid lava, built up far above the earth’s surface, thus appear as portions of a system of self-act- ing machinery by which the comparatively cold crust of the earth is sustained upon its less dense, and white-hot, molten sub- stratum. : As many geologists refer volcanic action mainly to the occlusion of water or steam in the lavas, rising from great depths, 1t 1s a most important point to ascertain whether or not the Hawaiian lavas do, as a rule, occlude water or steam. Captain Dutton —though he does not attribute the main action to this cause—seems to have no doubt of it, and refers to any fragment of Hawaiian pahoehoe as a proof. He might have added aa, for both are constantly crowded with vescicles. But what is the proof that these vescicles are caused by steam or vapor of water? In ny view there is nosuch proof. Ihave always looked upon these vescicles—wherever found, except 1n certain very exceptional cases—as simply air bubbles. The most notable instance of this vescicular structure is to be ob- served at all our great lava fountains. The lava spouts up under a head of liquid and forms a fountain, and at the same time a lake or pool of molten lava 1s formed around it. A fountain of any liquid thus thrown into the air and falling into the same liquid below produces a mass of foam or air bubbles which are again thrown up into the air. Lava foam when thus thrown up is instantly converted into glass-foam, which is the simple origin of our so-called pumice, it being a congeries of glass air- cells so light as to float in water. When the light frothy mass of molten glass runs away on the river of molten Java which proceeds from these fountains, it cools rapidly and forms the clinkery sur face of our aa streams. The interior of an aa stream, when it ‘has had time to settle and quietly cool, is often quite compact, without vescicles. But even the interior portion which pours in little cataracts over the uneven surface of the ground, must be constantly enclosing air bubbles which get frozen in. Let any one watch even a gen- tle stream of water pouring from a small height into a reservoir, he will find the water in the track of the entering stream to be full of air bubbles, a fountain or cat- aract produces a mass of foam. I am well aware that tbe freezing pro- cess tends to exclude air bubbles and even solids in solution, and this effect is com- monly seen in our pahoehoe lavas. The outer thin crust is often composed of a compact glass without vescicles; the freez- ing process, however, which of course, commences from the outside has frozen the bubbles in, for just inside the compact glassy coating, perhaps half an inch or more thick, the air bubbles are closely com- pacted. At the lava lakes of Kilauea other pro- cesses are constantly going on to fill the molten lavas with air bubbles whigh get frozen in in the same manner. For in- stance, as I interpret the facts—which slightly differs from Captain Dutton’s ex- planation—the cooled upper layers of lava descending, and the hotter layers from be- low rising, cause violent convection cur: rents and even whirlpools, which draw down the air with them. This air imme- diately gets intensely heated in the white- hot lava, becomes in consequence a highly elastic gas, and flies back, bringing with it jets, sprays and clots of lava, as well as Peles hair and pumice or glass foam. Heated air, is, on this view of the facts, the main elastic gas which is seen risin with the molten lavas in the lakes of Ki- lauea, although immengp quantities of steam or other dense white vapors are of- ten seen to arise from cracks in the neigh- borhood of the lakes, and when it rains— which is not seldom—the steam is some- times blinding. On a fine day I have how- ever, watched the active molten lakes for hours, without seeing a particle of visible steam proceeding from them or even from the neighborhood. : , Again, the blowing cones in the neigh- borhood of the lakes, which give intermit- tent reports, sometimes accompanied by jets of lava, seem to be another phase of the action of compressed and heated air. They are generally situated over caverns in which the molten lava—protected from radiation and cooling—may be seen swash- ing about like the waters in the rocky caves of our sea coast. When the level of the lava rises in the red-hot caverns and entraps air in the hollows of the roof, the heat rapidly expands it, and it soon bursts through the molten lava with an explosion carrying some of it through the opening which in falling forms a cone of glassy s.ag around it. The lava thus blown out is soon supplied again from the lake and when it has risen sufficiently the process is repeated. All the noisiest explosions which I have heard on Hawaii have been undoubtedly produced by compressed and heated air. With regard to the statement so often made respecting steam escaping from Ha- waiian molten lavas, I have a letter by me from the late Mr. Coan, of Hilo—who per- haps saw more running lava in his lifetime than any other man ever did—in answer to my inquiries on the subject, in which he states his belief that steam is never seen coming from Hawaiian lavas except when they .clearly come in contact with water, and then the steam comes from the water, not from the lava. From the orifices of eruption, he says he has often seen steam, and then he observes: “We may be sure that fire and water are in contact below.” «lhe late Mr. Poulet Scrope, in his work on volcanoes, made a serious mistake in quot- ing Mr. Coan in aid of the mixed steam and lava hypothesis, in which he makes him say, that “from the surface of the lava current of the eruption of 1852, the steam rose in fleecy wreaths towards heaven.” Mr. Coan was only describing the rise of great volumes of steam from the pools of water as the lava poured into them in its course, and which process was to be seen almost daily near Hilo in 1881. Professor Dana refers to the hanging cloud constantly seen over Kilauea, as a proof that steam escapes in large quanti- ties from the Kilauea lavas, and that this cloud often condenses and falls in rain. Much steam from percolating water no doubt arises from fissures all about Kilauea and when it rains ‘hard, from all the hot places, and from the surface of the lakes, and it usually blows away very quickly. But I attribute the hanging cloud to a dif- ferent cause—simply to that of all hanging clouds—the heated column of air over Ki- lauea absorbs and retains an excess of moisture in an invisible state, it rises by its less specific gravity into a cooler stza- tum and there condenses. The hanging clouds on the windward side of all cur mountains are merely another phase of the same principle. I have two oil paintings executed by Mr. Furneaux, of a remark- b able cloud which hung for a long time over that part of the 1881 flow, to the eastward of Hilo where the lava had got dammed up at the edge of the woods aud remained for a long time as a large half-raolten lake. This cloud not only dropped rain, but one of the pictures shows a water-spout pro- ceeding from it, and when the lower end of it touched the hot lava it was sent his- sing off again in steam. Now [would ap- peal to any observer of the ieva flow of 1880-1881-—and they were numerous— whether in this situation and away from any cracks, fissures, or orifices of eruption, there was any such amount of steam pro- ceeding from the lava, or from any orifices in that neighborhood, as could have pro- duced that constant hanging cloud, highly charged with vapor. A rising column however, of moist Hilo atmosphare heated over the molten lake and therefore contin- ually rising into a cooler stratura must pro- duce a hanging cloud, by the condensa- tion of the excess of moisture held by the heated air. It may well have produced water-spouts, for such cloud would soon become super-saturated and tend to fallin a vortex ring. This cloud was no doubt assisted, blackened and intensified by the smoke and vapors from the burning woods around the edges of the constantly ad- vancing lava lake. For my part I closely watchbd the lava flow of 1859, and walked upon the aa bank as the molten river ran down the center, I saw it afterwards fall into the sea at the coast. I have watched the lavas in the lakes of Kilauea, in the fountain of 1873 in the summit cra- ter, and in the flow of 1881 near Hilo, and I have never been able to detect steam coming from the lava, although I have looked for it. Steam from the ocean and from the pools of water that the lava ran - into, from the orifices of eruption, from cracks and fissures in the neighborhood of the hot rocks near Kilauea and else- where, from the trees and the jungle from which the lava was evaporating the juices from the rain that has poured on to it, 1 have seen in abundance, but coming from the lava itself I have never Leppened to detect it. Perhaps others havo been more fortunate. Why is it necessary, however, to imagine the cells in our lavas to be the result of steam, when air gives such a sim- ple account of them ? But there is still another phase in the behavior of our lavas, for which steam is constantly brought forward as the origin, and Capt. Dutton seems inclined to follow the usual interpretation—I mean the hol- low mounds of paheehoe, which some- times cover, in close succession, hundreds, perhaps thousands of acres of ground. He observes: “Tho numberless mounds or bosses of pahoehoe were all formed in detail in the manner already described, by repeated outshoots of streamlets from beneath the hardened crust behind. As these belches of lava cool they exclude the occluded steam, and the mass swells up by the formation of myriads of vescicles, and often by the formation of great hollow blisters underneath.” Having watched the formation of these mounds of pakoechoe in 1859 and in 1881, it did not appear to me that steam was con- cerned in the process. They seemed to be a simple hydrostatic effect in a liquid of which the crust cools rapidly. A vis- cous liquid usually tends to flow, if some- what slowly released, in flattened spher- oidal masses, as porridge does, or on a smaller scale like quicksilver. The rapid cooling and stiffening of the crust, fixes, as it were, these flattened spheroidal masses, and the molten lava flowing from the aa bank, or from a somewhat higher level behing, and protected by the crust trom radiation, pours into the rounded mass under a small head of pressure, and swells it out while the crust is somewhat plastic. But now two things occur, the crust gets stiffer and the head of ligtud inside gets higher. A break has to occur somewhere under the head of liquid pres- sure. This naturally happens at the low- est side, the molten lava belches out again and forms another mound alongside but a little lower down the slope. And so the process goes on. The molten lava often runs all out of the mound, leaving the hardened hollow dome, just as it runs out of the arched stream, leaving the crust with a tunnel underneath. Sometimes the contraction of the cool crust of the domes, causes the center piece of the top to crack in a rudely pentagonal or hexag- onal shape and to fall in, but sometimes it remains entire. In my view, if there were no water, steam, air or elastic gases in the planet, these domes would form essentially as they appear, while the Ha- waiian eruptions and fountains of lava might continue to spout as they do to-day. The whole seem to be simple hydrostatic effects, the result of liquids endeavoring 6 to find the level due to their density. There is this peculiarity about these li- quids, however, namely, that the slightest cooling makes them first viscid, then sol- id, so that they do not always respond as freely to changes of level as ordinary li- quids do, especially in narrow and crooked passages, or when exposed to radiation. From this cause, also, they exhibit the peculiarity of running up hill in certain circumstances. That is to say, the rapid- ly cooling crust forms a pipe in which the liquid under a head of pressure from be- hind ascends a slight slope in front of it, thus sometimes crossing hollows on the principle by which we see the water con- veyed in pipes across gulches in the course of the Spreckels ditch on Maui. On some- what similar principles, that is, liquid pressure from behind, and rapid cooling and solidifying outside, it is well known that a small, slowly advancing pakoehoe stream will sometimes cross on the sur- face of a narrow stream or pool of water, and when the water lowers, a natural bridge will be left standing. Mr. Fur- neaux has a picture of one of these naturs: - bridges so formed. Any one may have seen the lava thus advancing a short dis- tance over the surface of the pools near Hilo in 1881. The arched form of the crust over the lava tunnels on Hawaii has also sometimes been attributed without question to ex- panding steam or gases here is, how- ever, one evident reason for it, namely, that the middle of the molten stream is hotter than the sides, and therefore more expanded. But, besides this there is a well known principle which causes even slowly moving rivers to be higher towards the center than at the sides, that is, the increased velocity of the stream in the center and deeper parts, causes the liquid to stand higher, because the extra velocity of descent reduces the relative weight, so that it takes a higher column of liquid to} wards the center to balance a shorter one near the sides. The principle may be more easily understood, if we consider that if we could from above the surface of the earth, lower a spring balance with, say a ten pound weiglit on it, at just the ve- locity which a falling body acquires, the index would show ot zero; in other words we relieve the spring of the weight. A rapidly moving lava stream has all the features exaggerated which exhibit them- selves 1n a river. It would be difficult to do justice in a ~ | ry \ - 2 7 short notice, to all the different interest- ing points connected with Hawaiian vol- canoes to be found in Capt. Dutton’s re port on them. I have endeavored more particularly to call attention to a few of the problems which seem to admit of other solutions than those there presented, as well as to extend to volcanoes generally, Capt. Dutton’s view—which I have long advocated—that the surface of the lava in the Kilauea and Mauna Loa conduits, appear as equilibrium levels depending upon their mean density at any given time, and so as to be applicable to the gen- eral hypothesis of volcanic action, assum- ing a thin crust of the earth and a molten substratum. If this less density of the upper layer of the molten substratum—when at least it expands into a fissure in the crust— ‘may be fairly ‘inferred as probable, we should have a ready means of accounting for the different phases of volcanic action whether hydrostatic or explosive, in an- cient as well as in modern periods, espe- : wmiolly when we consider them in connec- "tion, with Elie de Beaumont’s important principle of the constant struggle between ‘he deformation of the spheroid by internal loss of heat, and the tendency of the earth’s rotation to maintain that spheroidal figure more nearly exact. gi Accidentally occluded water or steam, which seems to be so common in European volcanoes, might then—like geysers—take its proper position as a supplementary or parasitic phase of volcanic action, while an intelligible cause would present itself for the behavior of certain active volca- noes, not Hawaiian alone, but such as Charles Darwin refers to in his account of those in the Chilean Cordillera, situated far apart from each other, and built up . “in great snow-clad cones,” far above the earth’s mean surface; and the lavas in .which simultaneously appearing during the great regional earthquake of 1835, he describes as reminding him of “water splashing up through holes in the ice of a frozen pool when a person stamps on the surface.” Captain Dutton refers in the concluding paragraph of the chapter on the Volcanic Problem to “one characteristic of volcanic action, which is quite universal,” that is, “the intermittent feature of eruptions,” by which, “volcanoes do not discharge all their available product at once, but by re- peated spasms of activity, separated by intervals of repose,” the cause of which he seeks in some unknown “elastic force,” which gradually develops and is then dis- charged. “The agency,’ he says, “which thus progressively develops the potential energy or elastic force, is the missing fac- tor. When we discover it we shall discover the secret of the volcano.” As far as the eruptions of Hawaiian volcanoes are concerned however, Captain Dutton seems to have sufficiently explained their inter- mittent action in the second paragraph quoted in the commencement of this no- tice,for he looks upon the columns of lava in Mauaa Loa and Kilauea as steadily rising columns, while an eruption only means, that the liquid columns discharge them- selves at a lower level, when their height exceeds the retaining limits of the neigh- boring crust. The columns then take some years to reach or exceed the old levels again, when the process is repeated. In this way the eruptions appear intermittent, and the cones are gradually built up by a succession of lava streams, the lava some- times, however, being discharged at the summit instead of the sides, when the lat- ter are strong enough to resist the hydro- static pressure. The “missing factor” therefore, does not seem, in these cases, to be connected with the intermittent ac- tion, nor necessarily with any accumulat- ing elastic force, but rather with the power which has kept these lavas slowly and steadily rising through the ages, as far back at least as “the life of Mauna Loa.” Some phase of the secular deformation of the spheroid—a process which we in the Pacific Ocean see in action all around us in the slow subsidence of its bed, not- withstanding the local upheaval of active volcanic areas—may perhaps indicate one means at least, by which the molten sub- stratum may be gradually pressed up through fissures in the crust, while the normal level of the liquid upper layer when expanded into such fissure, and leav- ing the heavier components below, may be many thousand feet above the mean sur- face of the earth’s crust, owing to its less density. However this may be, Hawaiian volcanoes, in the grander phenomena as well as in the details of their action, seem to be operated by hydrostatic rather than by elastic forces, although the latter may well be every now and then exhibited, as a noisy and demonstrative, but really sec- ondary feature. Honolulu, August 25th, 1884. OF TITLE END OF REEL. PLEASE ~ REWIND.