order as the explosions which in recent years havebeen witnessed at Krakatoa , in the Straits ofSunda, and at the volcano of Bandaisan, in Japan .There is great similarity in composition and instructure between the bedded series under dis-cussion and certain members of the volcanic com-plex of the Yellowstone National Park , but in thelatter case evidence of slow accumulation is foundin the succession of fossil forests destroyed andburied by the andesitic tuffs and breccias.

The Intermediate series. —The alternating seriesof andesitic and rhyolitic lavas included in thisgroup serves to show that at a certain time achange took place in the character of the volcanicproducts. Lavas of rhyolitic composition wereerupted for the first time, but alternated withandesitic outpourings in a manner to suggest thatdifferent vents must have been emitting differentmaterials at the same time.

The Intermediate series has a much more varia-ble development than the San Juan, as now seen.It is thickest near Ophir Pass, and extends as farwest as the Lizard Head.

The Potosi rhyolite series. —The uppermostmember of the volcanic complex in this quadran-gle is composed almost wholly of rhyolite, andindicates an important epoch, in which andesiticeruptions had practically ceased. There is nomeans of knowing how fully the flows Rhyolite rep ,and tuffs seen in the Telluride quad- a e dfs n ttnct ve °*rangle represent the whole series of epoch 'eruptions in this epoch, but it must be assumedthat there were still higher flows or tuff layers,now completely removed from the highest peaksof the region.

As with the lower members of the localsequence, the Potosi flows and tuffs are so nearlyhorizontal and cover so much space that no infer-ence of value can be drawn as to their source orextent in other parts of the San Juan. Hikes ofglassy rhyolite are said to occur in Potosi Peak,but that these fissures were the channels of erup-tion for the thick flows of the same mountaindoes not seem possible.

Dike eruptions. —The three bedded members ofthe volcanic succession are cut by narrow dikes ofpyroxene-andesite, especially in the northeasternportion of the quadrangle. These dikes cut everyother rock in their paths, but thereis no certain evidence that the andes- ing to a late^itic dikes belong to the same period oferuption as the basic dikes associated with thediorite-monzonite stock of the Mount Wilsongroup. From the uniform character of the andes-ite dikes in the eastern portion of the quadrangleit seems probable that they belong to some lateepoch of activity in the cycle of the San Juanvolcano not otherwise represented in this area.

Period of waning volcanic energy. —When theentire San Juan region has been studied manyphenomena will undoubtedly be observed whichmust be explained as belonging to the laterphases of activity commonly noted in volcanic dis-tricts, such as local eruptions of peculiar lavas,solfataric and fumarole action, hot springs, etc.The Telluride quadrangle, however, seems suffi-ciently removed from the real centers of volcanicaction to make it a matter of some doubt as tohow far the agencies which have producedchanges in the rocks, or have deposited secondaryminerals in them, may be considered truly volcanicagencies. Decomposition of the fragmental andof some of the massive rocks has taken place,but this would be a natural result of the percola-tion of surface waters.

LACCOLITHIC INTRUSIONS.

the stock eruptions is based upon the relation ofsheets and stocks of similar rocks in the Rico andLa Plata mountains. In both of these districtsdiorite- and monzonite-porphyries are cut by dio-rite stocks. In the Telluride quadrangle no evi-dence bearing upon this point was observed.

STOCK ERUPTIONS.

The large stocks of the Telluride district pre-sent a number of interesting problems. The formof the masses and their clearly exposedrelations to the sedimentary formations other'erup=and the bedded volcanic series showthat the stocks now seen represent the filling ofimmense conduits which penetrated to the highestlevels now existing in this part of the San Juan,and which, it must be assumed, ascended to thesurface of the time of eruption. If they didextend to the surface, lavas must have issued fromthem, and they are thus volcanic channels, but noevidence has been found indicating that they arethe throats of explosive volcanoes, such as thosefrom which the great fragmental masses of theSan Juan were derived.

The rocks of these stocks are mainly granularin structure and are often rather coarse grained,as in the case of the Stony Mountain gabbro. Ithas been a belief among petrologists that suchstructures could result from the consolidationof igneous magmas only at great depths — depthsmeasured by at least several thousand feet. Toassume that the rocks now seen in the .

Importance

summits of Mount Wilson, Mount Snef- l^ s h? s to?y C of sfels, and other high peaks were consoli- San Juan 'dated at even 3000 or 4000 feet below the surfaceis to add to the volcanic complex of the San Juanan enormous amount of material. But unless itis supposed that these granular rocks were formedvery near the old surface they must be consideredas giving clear evidence of a former extent of thevolcanic pile of the San Juan, beside which itspresent dimensions become insignificant; andwhether the visible monzonites, diorites, or gab-bros formed at 1000 or 10,000 feet below thesurface, it is plain that the typical porphyries ofthe region, including the Rico and La Platamountains, consolidated at the same or greaterdepths; so that the belief, finding most positiveexpression in the German school of petrography,that great depth is essential to the formation ofgranular structure in large igneous masses, whilethe porphyritic structure belongs to higher zonesin the earth’s crust, is clearly contrary to the factshere revealed.

The petrographic complexity of some of thesestocks is worthy of much more detailed examina-tion than could be given it in this work. Theone exhibiting the greatest variation is that of theOphir Needles, while a considerable diversity ofrock varieties was also noted in Mount Sneffelsand in Mount Wilson. Most of the changes incomposition appear gradual and are not explain-able as due to distinct eruptions of differentmagmas. In some cases sharp contacts werefound. No regular relation between the variationand the form of the stock was observed. Somestocks, as that of Grizzly Peak and RollingMountain, are nearly homogeneous in mineralcomposition throughout. No fragmental materiallike agglomerate occurs in these stocks, and theyare not centers from which dikes radiate. Thesefacts do not allow of the supposition thatrepeated eruptions of different magmas tookplace in any one of these conduits to a degreewhich can permit the idea that they are possiblythroats of typical volcanoes.

The laccoliths of the Telluride quadrangle,together with the small sheets and dikes nearthem and the granite-porphyry mass of Relation ofHoward Fork, are thought to have been {“vSlSSficintruded after the surface eruptions ofthe bedded volcanics, but before the stock intru-sions. The granite-porphyry mass penetratingthe San Juan tuffs affords direct evidence of thisfact, and the presence of Mancos shales on thesummit of Whipple Mountain seems to give infer-ential testimony in the same direction. It doesnot seem probable that these shales would havebeen preserved at this point, 600 or 700 feet abovethe plain of erosion upon which the San Miguelbeds rest in Ruffner Mountain, if the planation ofthe region took place after the intrusion of thediorite-porphyry.

The inference that the laccoliths are older than

RELATIONS BETWEEN THREE TYPES OPERUPTION.

Three very different phases of eruptive activityseem to be illustrated in the Telluride quadrangle.The products of eruption do not vary ProbIemsastoessentially in chemical composition; at i g neousrocks -least some of the stock diorites, laccolithic por-phyries, and surface andesites correspond closelyin chemical composition. But if all are derivedfrom the same source of molten material, as mightbe inferred from this chemical composition, itappears that the eruptions took place under dif-ferent physical conditions; that the forces impel-ling the eruptions were of different kinds, or ofgreatly varying intensity if of the same kind. Itis not intended to discuss this problem at lengthin this place, for it is hoped that light may beshed upon the connection between these several

types of eruption through the investigations ofother portions of the San Juan. But the formwhich the problem takes as presented in thisquadrangle will be briefly stated.

The question may be expressed in general termsas follows: Were the magmas of the intrusivelaccoliths, the stocks, and the effusive flows ofthe Telluride quadrangle derived from the samesubterranean source and impelled to or towaidthe surface by volcanic forces of the same origin,or were they derived from different levels andimpelled by forces differing somewhat in kind?The laccoliths are similar to thousands of por-phyry masses scattered through Colorado and theadjacent plateau country of Utah, Arizona , andNew Mexico , in regions where no evidence nowremaining suggests that surface volcanic eruptionsever took place. The stocks of this region areconduits, so near the old surface that they seemto represent channels through which lavas werequietly poured out. But exactly similar stocksare present in the Rico, La Plata, and Elk moun-tains, and probably in other mountain groups ofthe Great Plateau. It thus becomes questionablewhether the association of these intrusive masseswith the surface volcanics of the San Juan isevidence that they are products of eruptive activ-ity about one great center, or whether they arereally independent of the typical volcanic mani-festations of the San Juan.

TERTIARY AND RECENT OROGRAPHICMOVEMENTS.

The general problem. —The area of the SanJuan Mountains seems to have been a land massfrom the time that the San Miguel lake 0rographicwas filled up by the great series of vol- Jlulny™*-

. t i termined.

came rocks. As no sedimentary bedsexist by which the earth movements of Tertiarytime can be differentiated within the mountaindistrict, a detailed study of the whole region isnecessary before any definite statements can bemade as to the various movements of uplift orsubsidence by which the San Juan district hasbeen affected. That the area is now one of themost elevated in the United States , and that avast amount of denudation has already beenaccomplished, are the two great facts which testifyto the magnitude of the combined disturbancessince the beginning of the volcanic eruptions.The San Juan area has, moreover, been upliftedas a great continental mass, and not by axial foldslike those of some of the Rocky Mountain rangesof Colorado .

The Hayden map shows a lake-bed deposit inthe valley of the Rio Grande above Wagon WheelGap, bearing the symbol of the Green RiverEocene , but the reports give no information as tothe grounds for assigning the beds to that age. Onthe southern flanks of the San Juan the Puercoand Wasatch divisions of the Eocene are presentin the drainage area of the San Juan River, andthey are to some extent upturned nearest themountains. Since they do not come in contactwith the volcanic formations of the San Juan theinfluence upon the latter of the movement affect-ing these Eocene deposits can not be determined.

The discussions of orographic movements inthe Rocky Mountains by S. F. Emmons (BulletinGeological Society of America, Vol. I, 1890, pp.245-286) and R. C. Hills, (Proceedings ColoradoScientific Society, Vol. Ill, 1890, pp. 362-458)present valuable generalizations as to the succes-sion of movements recognized at various places,but in their application to the San Juan mountainregion proper the conclusions are necessarilylargely speculative. The unconformity at thebase of the San Miguel shows that much of theupturning of the Mesozoic beds adjacent to themountains, which, as seen on the Animas Rivermight be considered as the same movement thataffects the Puerco, is really in large degree post-Cretaceous , unless the San Miguel and the vol-canic series are all of much later date than is nowbelieved.

Tilting of the volcanic complex. —In the Tel-luride quadrangle the two definite structuralphenomena referable to Tertiary move-ments are the tilting of the whole movementbedded volcanic series, with the San recognized *Miguel at their base, and the faulting by whichall formations seem to be affected. In the mainvalley of the San Miguel the conglomerate of thatname descends more than 1000 feet in the distanceof 8 miles between Iron Mountain and IngramCreek. This is due to a gentle eastward inclina-

tion of the strata. In the central portion of thequadrangle a northeasterly dip of a very fewdegrees prevails, with many local irregularitiesdue to the large igneous intrusions. In the south-eastern corner of the quadrangle the base of theSan Miguel is 2000 feet higher than at IngramCreek, and the strike is nearly east and west.Local undulations, such as those in the Twin Sis-ters and in the ridge south of Rolling Mountain,are found here and there.

This general northeasterly dip of the SanMiguel and overlying volcanic formations has, nodoubt, had an important and perhaps predominantinfluence in determining the broad features of thedissection of the San Juan volcanic plateau byerosion, at least in its western portion.

Faults and fissure systems. —The faults observedto cut the volcanic series are of small dislocation,trend in various directions, and display Many _no system. Some of them are ore-bear- *“ r t e r s e ^^l 2 .ing veins, and possibly some of the large ab,e system 'veins upon which there is no notable displacementwhere exposed may belong to the same period ofAssuring with certain of the faults.

In the description of the economic features (p.15), by Mr. Purington, will be found statementsas to the several systems of joints or fissureswhich are locally prominent in the Telluridequadrangle, and in his full report Mr. Puringtondwells at some length upon the origin of thesefractures. As the district studied is but a smallportion of the San Juan region, no conclusivestatement can as yet be made regarding the rela-tion of these fissure systems to broad movementsin the San Juan, nor can the fissures be classifiedin sets of contemporaneous origin. That some ofthese systems of fissures have been of much localimportance in determining the course of erosioncan not be questioned.

DENUDATION OF THE SAN JUAN PLATEAU.

Character of the plateau. —From present knowl-edge it appears that at the period of maximumdevelopment the accumulations fromthe volcanic centers of the San Juan tent of San

. „ , . Juan Plateau.

must have formed a great plateau,much greater in extent than the area now coveredby the lavas and tuffs. The Telluride quadrangle,now situated on the extreme western border of thevolcanic complex, must once have been far withinits limits, and we have here the evidence of enor-mous erosion, by which the abrupt western frontof the San Juan Mountains has been carved outof the old plateau. Degradation of the volcanicpile by ordinary agencies of subaerial erosion hasdoubtless been in progress since early Eocene time, the work of many of the earlier intervals inthe volcanic history being undone by succeedingeruptions.

The bedded series of tuffs, agglomerates, andlava flows, still preserved in the higher ridgesand peaks of the Telluride quadrangle, provesbeyond a doubt that a thickness of 3000 or 4000

feet of these rocks once covered the entire quad-rangle above the level of the San Miguel con-glomerate. The evidence of the large stocksindicates that a considerable further thickness ofsimilar volcanic materials has been entirelyremoved, so that it seems to the writer quitewithin the bounds of reason to assume that theremay once have been 6000 or 7000 feet of volcanicrocks present in this region. How far the vol-canics extended to the westward is totallyunknown, but it must have been many miles,judging from the thickness still preserved in thisquadrangle.

The cause of denudation. —The work of erosionon this, the western, side of the San Juan hasresulted in the almost complete removal c .o± the entire bedded series back to the XTcSL-abrupt front which has been described. pared "

But erosion has not stopped with removal of thevolcanics. The sedimentary rocks have beendenuded over large areas down to a new plateaulevel that of the Dakota sandstone — and deepcanyons have been carved in the harder rocksbelow it. In view of the eastward tilting of theregion, it is probable that the San Miguel Riverat Sawpit, only 16 miles from its present head,has cut at least 7000 feet, and possibly more than10,000 feet, below the surface which existed atthe close of volcanic activity. The penetrationof the river to-day is only 8 miles beyond themountain front. The Dolores River has also

accomplished a great task of removal.