&, 
 
 UC-NRLF 
 
 B H 17S 7Sfl 
 
 M01J& OJ! 1 TMJfi GEOLOGICAL SURTET 
 
 ENGLAND AND WALES. 
 
 THE FELSITIC LAVAS 
 
 OF 
 
 ENGLAND AND WALES, 
 
 WITH AN 
 
 INTRODUCTORY DESCRIPTION OP THE CHIEF 
 CHARACTERS OE THIS GROUP OP ROCKS. 
 
 BY 
 
 FRANK RTJTLEY, F.G.S., 
 
 LECTURER ON MINERALOGY, ROYAL SCHOOL OF MINES. 
 PUBLISHED BY OBDER OP THE LORDS COMMISSIONERS OF HEB MAJESTY'S TBHAIUBY. 
 
 LONDON: 
 
 PRINTED FOR HER MAJESTY'S STATIONERY OFFICE j 
 
 AND SOLD BY 
 
 LONGMANS & Co., Paternoster Row; TRUBNER & Co., Ludgate Hill; 
 LE i v Co., Limited, 33, King William Street, E.C. ; 
 
 E. STANFORD, 55, Charing Cross; WYLD, 12, Charing Cross; 
 
 and B. QUARITCII, 15, Piccadilly. 
 
 Manchester : T. J. DAY, 53, Market Street; 
 
 Edinburgh : W. and A. K. JOHNSTON. 4, St. Andrew Square ; 
 
 Dublin : HODGES, FIGGIS, & Co., 104, Grafton Street ; and 
 
 A. THOM & Co., Limited, Abbey Street. 
 
 1885. 
 
 Price JS'inepence. 
 
/^BERKELEY 
 
 LIBRARY 
 
 UNIVERSITY OF 
 V CALIFORNIA 
 
 EARTH 
 
 SCIENCES 
 
 LIBRARY 
 
 EXCHANGE 
 

MEMOIRS OF THE GEOLOGICAL SURYET. 
 
 ENGLAND AND WALES. 
 
 THE FELSITIC LAVAS 
 
 OF 
 
 ENGLAND AND WALES, 
 
 WITH AN 
 
 INTRODUCTORY DESCRIPTION OF THE CHIEF 
 CHARACTERS OF THIS GROUP OF ROCKS. 
 
 BY 
 
 FRANK RUTLEY, F.G.S., 
 
 LECTURER ON MINERALOGY, ROYAL SCHOOL OF MINES. 
 
 PUBL1SFIED BT ORDER OP THE LORDS COMMISSIONERS OF HER MAJESTY'S TREASURY. 
 
 LONDON: 
 PRINTED FOE HER MAJESTY'S STATIONERY OFFICE: 
 
 AND SOLD BY 
 
 LONGMANS & Co., Paternoster Jiow; TRUBNER & Co., Ludgate Hill ; 
 LETTS, SON, & Co., Limited. 33, King William Street, E.C. ; 
 E. STANFORD, 55, Charing Cross; WYLD, 12, Charing Cross ; 
 
 and B. QUARITCH, 15, Piccadilly. 
 
 Manchester : T. J. DAY, 53, Market Street; 
 
 Edinburgh : W. and A. K. JOHNSTON. 4, St. Andrew Square ; 
 
 Dublin : HODGES, FIGGIS, Co., 104, Grafton Street ; and 
 
 A. THOM & Co., Limited, Atbey Street. 
 
 1885. 
 
 Price Ninepence. 
 
EARTH 
 
 SCIENCES 
 
 LIBRARY 
 
 *** 
 
CONTENTS. 
 
 Page. 
 Preface. By the Director-General. 
 
 Introduction ........ i 
 
 Felsitic Lavas of Shropshire, &c. - - -11 
 
 ,, ftie English Lake District - 12 
 
 North Wales - -15 
 
 .. Skomer Island - 18 
 
 ,, ., Pembrokeshire - - 20 
 
 Post-Silurian Felsitic Lavas - - - - 22 
 
 Felsites of Doubtful Origin - - - 
 
 Tabular Statement - ^ - 2o 
 
 v -. 
 
 Alphabetical List of Localities - - -27 
 
 Explanation of the Plates (I. to IT.) - - - 28 
 
 u 18642. Wt. 8917. A 2 
 
 313025 
 
PREFACE. 
 
 IN the following pages a summary is given of all that is at pre- 
 sent known regarding the distribution of that interesting group 
 of rocks, the Felsitic Lavas, in England and Wales. Mr. Rutley 
 has incorporated the published observations of other petrographers 
 with those made by himself in the Collections of the Geological 
 Survey. As the number of instances of devitrified rocks is con- 
 stantly being increased in the British Islands and abroad, and a& 
 the subject of devitrification is one on which much light still 
 requires to be thrown, both by observations on ancient rocks and 
 by laboratory experiments, it is believed that the publication of 
 this Memoir will be of service. 
 
 ARCH. GEIKIE, 
 
 Director-G eneral. 
 
 Geological Survey Office, 
 
 28, Jermyn Street, S.W., 
 21st November 1885. 
 
THE FELSITIC LAVAS 
 
 OF 
 
 ENGLAND AND WALES, 
 
 WITH AN 
 
 INTRODUCTORY DESCRIPTION OP THE CHIEF 
 CHARACTERS OP THIS GROUP OP ROCKS. 
 
 INTRODUCTION. 
 
 Ten years ago the former occurrence of vitreous rocks in 
 England and Wales was not demonstrated, even if suspected. 
 The terms felstone, felsite, petrosilex, hornstone, eurite, &c., had 
 'for many years been applied to rocks of more or less compact 
 texture, often with conchoidal fracture, having approximately the 
 hardness and fusibility of felspars, and believed to be intimate 
 admixtures of felspar and quartz. Such rocks frequently 
 exhibited, especially upon weathered surfaces, a finely banded 
 structure, closely resembling the lamination of sedimentary rocks, 
 and where, as sometimes happened, they were found to be inter- 
 bedded with ancient stratified deposits, occasionally fossiliferous, 
 it is not surprising that they were now and then described as 
 " altered flags " and slates, to which, when in advanced stages of 
 metamorphism, these felsitic rocks certainly bear a somewhat close 
 resemblance. Their mineralogical affinity to the groundmass of 
 quartz-porphyries or quartz-fel sites and also to trachytes was 
 recognised. Many of them were mapped by the Geological 
 Survey as lavas, and more than twenty years ago, the banding in 
 one of the Welsh felsites was stated by Sir Andrew Ramsay to 
 have " probably originated in the same cause that produced the 
 lamination in the lava of Ascension." 
 
 Although a very large proportion of the English and W r elsh 
 felsites are now known to be true lavas, it should also be remem 
 bered that many rocks of quite different origin exist which, from 
 their mineral constitution and structure, may also be termed felsites 
 or felstones. To describe all the rocks which might be included 
 in the group of felstones would be a somewhat difficult and arduous 
 
2 > * . .. ... ,., . .INTRODUCTION. 
 
 task, even if the needful information were at hand ; although it 
 may, perhaps, be useful to give, in conclusion, a few selected 
 examples of felstones which have apparently had an origin 
 different from that of the lavas which specially constitute the 
 subject of these pages.* 
 
 It has already been stated that felsites are believed to consist of 
 felspar and quartz. The constituent grains of these minerals are 
 very minute in rocks of this class, but when we examine thin 
 sections of felsite under the microscope, polarised light being 
 employed, we notice that there is a considerable difference in the 
 size of the crystalline grains in different kinds of felsite ; we also 
 see that there is sometimes a marked difference in the distinctness 
 with which the boundaries of the grains are visible, for while in 
 some cases their outlines are quite sharply defined, so that the eye 
 can clearly recognise the exact area occupied by each grain in the 
 section, as in the groundmass of an ordinary quartz-felsite or 
 elvan, in others the grains seem to shade off one into another, so 
 that no distinct boundary can be assigned to any particular grain, 
 in which case the section presents a hazy or blotchy appearance, 
 while, furthermore, we may in some instances meet with so fine a 
 texture that the field of the microscope seems merely to be 
 studded with minute points of light even under moderately high 
 powers. It is evident, then, that we may have very different 
 " polarisation pictures " resulting from different varieties of felsite. 
 
 The first of the structures just mentioned, in which the con- 
 stituent grains are distinctly marked and the several quartz and 
 felspar grains are capable of recognition under the microscope, 
 is termed a " micro- crystalline structure." 
 
 The second and third, in which the boundaries of the crystalline 
 grains are inappreciable and the different mineral characters of the 
 grains cannot be determined, is known a<? a " crypto-crystalline 
 " structure." 
 
 Both the quartz and the felspar exhibit double refraction, but 
 between these doubly -refracting or anisotropic grains a substance 
 may sometimes be detected in felsites which does not exhibit any 
 double refraction ; it is inert upon the polarised beam ; it is 
 isotropic. Such isotropic matter may be perfectly homogeneous. 
 This is glass or the " glassy basis " of Rosenbusch.t In its truly 
 vitreous condition it remains absolutely dark during a complete 
 revolution of the stage between crossed nicols. 
 
 In other instances isotropic matter is present which is not 
 homogeneous and structureless, but seems to consist of diminutive 
 fibres, scales, and granules with some interstitial substance, This 
 devitrified substance is the " micro-felsitic basis '' of Rosenbusch. 
 
 * It may be of interest to mention that a straight line drawn from Dunstanborough 
 Castle, N.E. of Alnwick to Skomer Island, off the coast of Pembrokeshire, passes 
 through some of the principal English and Welsh localities where felsitic lavas of 
 Silurian age are met with. 
 
 f Mikroskopischc Physiographic der Massigen Gesteine, p. 71. Stuttgart, 1877. 
 
BANDED STRUCTURE. 
 
 Any of the conditions above mentioned may occur in felsites, 
 constituting what is understood by the general term "felsitic 
 
 matter." 
 
 Fel sites, however, frequently exhibit other structures, some of 
 which are very characteristic and indicative of the conditions 
 under which the rocks were formed. 
 
 Banded Structure. This structure is often met with in felsites 
 of different kinds and sometimes of different origin. It is fre- 
 quently seen in those felsites which were once vitreous, and in 
 almost* all such cases it may safely be inferred that the banding 
 existed prior to devitrification. In unaltered vitreous lavas, 
 erupted within comparatively recent limes, the banding may be 
 of various kinds. Its simplest form consists, perhaps, in slight 
 differences in the character of the glass itself, evinced only by 
 difference in colour. In the obsidian of Ascension, for instance, 
 when a thin section is examined by ordinary transmitted light, 
 dark brown bands may sometimes be seen traversing a pale 
 yellowish or colourless glass. Frequently these bands have no 
 sharply defined margins, but fade gradually into the adjacent glass 
 as though the latter had been merely dyed with some brownish 
 pigment. At other times such bands of coloured glass may have 
 very distinct boundaries, as in an obsidian from the Yellowstone 
 District, in which sharply defined bands of deep yellowish or 
 orange coloured glass traverse a glass of very much paler tint. 
 
 Such glass bands sometimes depolarise, owing to strain, and 
 between crossed nicols they extinguish in the direction of the 
 fibres, i.e., when the fibres are brought parallel to the principal 
 section of one of the nicols the light, which was previously 
 transmitted by the fibres when in other azimuths, now becomes 
 extinct. It is a common thing to find such bands densely crowded 
 with the small crystalline bodies termed microliths or crystallites, 
 some of which may be seen to exhibit double refraction while 
 others show none ; also with minute spherical, isotropic bodies 
 termed globulites. 
 
 Frequently, too, in unaltered glassy lavas, distinct bands are 
 visible which, when examined under high powers, are seen to 
 consist almost exclusively of streams of microliths, without any 
 appreciable difference between the glass of the band in which 
 they occur and that adjacent to it, the glass being colourless, 
 or its tint uniform throughout (Fig. 7, PL 1.) Such rnicrolithic 
 streams or bauds may be seen in some of the perlitic lavas of 
 Saxony. 
 
 It would seem unreasonable to expect to recognise banding, of 
 the kinds just described, in devitrified rocks, since one would assume 
 that the superinduced crystalline structure would obliterate all 
 evidence of bandiug shown by mere difference of tint in the glass, 
 unless there were some sensible difference in the chemical com- 
 position of the more or less deeply coloured, or of the coloured 
 and colourless bands. And, so far as the bands of microliths are 
 
4 BANDED STRUCTURE. 
 
 concerned, we can well imagine that, although, in common with 
 larger crystals, they would survive the changes brought about by 
 devitrification, yet we can hardly expect to recognise the in- 
 dividuality of these diminutive bodies when surrounded by and 
 intermixed with the products of a subsequent crystallisation. It 
 is quite possible, however, that differences in the state of tension 
 of the material in and about the microlithic bands may give rise 
 to marked diversity in the texture of the crystalline products of 
 devitrification. Great differences in crystalline texture may often 
 be noted between the bands of devitrified lavas and the parts 
 separating those bands, the latter, as a rule, exhibiting a much 
 less coarse grain than the bands themselves (Fig. 8, PJ. 1). 
 Indeed, some difference in grain is always to be found where this 
 laminated structure exists. 
 
 The bands may be either approximately parallel or sinuous, and 
 it is common to find that the presence of a porphyritic crystal 
 or a large spherule is accompanied by more or less deflection of 
 the adjacent banding. 
 
 This phenomenon, taken in conjunction with the axial ly parallel 
 arrangement of microliths in the bands, has given rise to the terms 
 tf fluidal structure," " fluxion structure," or " flowage " as it is 
 sometimes termed ; the streams of microliths and other bands 
 being assumed to sweep past and around these larger bodies as 
 the water of a river sweeps round an island (Fig. 7, PI. 1). 
 When structure of this kind is present in a felsite it is usual to 
 accept it as evidence, partially confirmatory, of former vitreous 
 character, but it should be borne in mind that similar phenomena 
 are met with not only in trachytes and other crystalline lavas, 
 but also in some dykes, and even in metamorphic rocks, as pointed 
 out most forcibly in the recent works of Professor A. Geikie* and 
 and Dr. J. Lehmannf ; while like appearances may be noted 
 in boulder-clays and other stratified deposits where large un- 
 yielding masses cause a synclinal deflection of the subjacent, and 
 an anticlinal one of the incumbent beds.J 
 
 Sometimes the banding in felsitic rocks, and also in their more 
 .modern glassy representatives, becomes not merely sinuous, but 
 contorted, and, in rare instances, contortion may reach even the 
 pitch of intricate convolution, in which case a " damascened " 
 structure results (Fig. 10, PL 1). 
 
 Bands composed of spherules are often met with both in glassy 
 and in devitrified lavas. In the latter case the spherules are 
 frequently the product of devitrification, and it is not unusual in 
 a devitrified, banded lava to find that the spherules which devitrify 
 
 * " Text Book of Geology," 2nd Edition, see ' Shearing of rocks,' pp. 289, 506, 
 531, &c. Fig. 246 in this volume shows the close resemblance in the structure of 
 a compressed Cambrian sandstone to that of a rhyolite. 
 
 " Untersuchungen iiber die Entstehung der Altkrystall, Schiefergesteine." 
 
 J In a paper by the late Rev. J. Clifton Ward (Quart. Journ. Geol. Soc., 
 Vol. XXXI., p. 410), an "altered ash" is described, in which a structure, sug- 
 gestive of fluxion, is attributed to an alteration of the rock. 
 
 "The Study of Rocks," 2nd Edition, p. 181. 
 
SPHERULITIC STRUCTURE. 5 
 
 the bands differ considerably in size from those which devitrify 
 the rest of the rock. Occasionally they are much smaller in the 
 bands than elsewhere. 
 
 The banded structure of felsites is frequently rendered evident 
 by superficial inequalities formed upon weathered surface?, the 
 rock being more or less deeply eroded in fine and approximately 
 parallel furrows. Since similar banding and erosion may be met 
 with in altered sedimentary rocks, which, when in the condition 
 of porcellanite, offer just as good conchoidal fractures as felsites, it 
 is dangerous to base any positive conclusions upon this class of 
 evidence. 
 
 Spherulitic Structure. The development of spherules con- 
 stitutes sometimes the first, at others the last change in a glassy 
 rock. In the first case it represents a structure preceding, or 
 coeval with the solidification of the lava, and it may be restricted 
 to the development of isolated spherules or of single belts of 
 spherules, which may be separate or may have coalesced in such 
 a manner that their sections appear as beaded strings (Fig. 11, 
 PL 1) ; or the union may have been so complete that no con- 
 strictions indicative of beads are visible, in which case they form 
 a band with straight and parallel edges (Fig. 12, PL 1). 
 Spherules may also become clustered together in great numbers, 
 forming bands many spherules thick (Fig. 9, PL 1), and oc- 
 casionally they group themselves around porphyritic crystals and 
 along cracks in the rock.* Sometimes the development of 
 spherules is so great at the period of solidification of a lava that 
 they constitute the entire rock (" Sphdrolithfels " of German 
 petrographers). In other cases a vitreous rock, which may possibly 
 have been quite free from spherules when it solidified, may have be- 
 come completely devitrified through the formation of these bodies, 
 as in the devitrified obsidian of Till's Hole (described on p. 12). 
 
 The spherulitic structure of a rock is sometimes rendered evident 
 through weathering, the weathered surfaces often looking as if 
 they were studded with whitened peas or shot, and occasionally 
 the spherules are so large that they may be likened to 
 tennis tails. It would be useless to attempt to classify these 
 bodies according to their size, for they vary from two or 
 three inches or more in diameter to absolutely microscopic 
 dimensions. In the case of spheroidal concretions or separations, 
 it may be a matter of feet rather than inches. 
 
 In a paper on artificial devitrification by Mr. D. Herman and 
 the authorf, instances are given in which pieces of plate glass, 
 submitted to a high temperature for a few hours onlv, have had 
 merely the surfaces slightly affected. Such surfaces show that 
 exceedingly minute spherical bodies (" globulites " of Vogelsang) 
 have been developed in great numbers, and, when the" film in 
 which they occur is examined microscopically under high powers, 
 
 * Quart. Journ. Geol. Soc., Vol. XL., p. 344. 
 
 f Proc. Royal Soc., Vol. XXXIX., pp. 87-107. Plates 1-4. 
 
6 SFHKRULITIC STRUCTURE. 
 
 they are seen to have massed themselves together in certain spots, 
 giving rise to little granular segregations (Fig, 6, PL 1). From 
 an examination of many specimens of glass, operated upon for 
 varying periods in a similar manner, some of which contain well- 
 developed spherules with a radiating crystalline structure, it seems 
 more than probable that the little groups of globulites represent 
 an embryonic stage in the formation of spherules. Vogelsang* 
 states that he has seen a crystal increase by the accretion and 
 transformation of globulites., and he has noticed, in the course of 
 his experiments, that globulites, so long as they remain isolated 
 in the magma in which they have been formed, remain isotropic, 
 but that, as soon as they become grouped together into crystal- 
 lites, double refraction usually becomes apparent and that this 
 property is only feebly developed when the crystallites- are 
 imperfect. There is, then, under such circumstances, a passage 
 from the amorphous to the crystalline condition, but when, through 
 increased rigidity, the resistance of the solidifying magma becomes 
 augmented, the globulites, even when in actual contact with a 
 crystal, cease to be assimilated by it, and permanently retain their 
 spherical form. 
 
 Globulites are 'sometimes seen to be densely grouped about 
 spherulites (as in Fig. 2. PI. 1) so as either to form a more or 
 less thick envelope around them, or in great part to constitute the 
 spherulite. Examples may be seen in the obsidians of Vulcano and 
 the Kocche Eosse, and it is interesting to note that this globulitic 
 matter depolarises in a manner indicative of a radial crystallisation. 
 A careful examination of many of these spherules shows that 
 there is, indeed, a divergent crystallisation set up within these 
 globulitic masses. When globulites come into contact and 
 arrange themselves in lines, as they frequently do, such linear 
 aggregates have been called " margarites," from their resemblance 
 to strings of minute pearls ; while, when the globulites in such 
 linear order became merged into one another, so that the globular 
 forms disappear, and minute, elongated, cylindrical bodies result, 
 the latter are termed " longulites." 
 
 It would seem, then, that in the cases just cited we have longulites 
 present, mixed with globulites which may or may not have 
 a linear arrangement, and it is probably to the presence of 
 divergent groups of longulites that the double refraction is 
 due ; since, although it might be difficult to prove the existence 
 of this property in a single longulite, yet it is possible that 
 bundles of them may exert an appreciable depolarising effect. 
 The same may be said of the depolarisation consequent upon 
 strain, which is visible in a bundle of fibres of spun glass, 
 while in a single fibre no depolarisation is apparent. If the 
 globulites be solid microscopic pellets of vitreous matter, 
 
 * " Die Krystalliten." Bonn, 1875, aud "Archives Neerlandaises," 1872 ; also 
 Zirkel's Mik. Beschatf. der Min. u. Gest, p. 95. 
 
SPHERULIT1C STRUCTURE. 7 
 
 as assumed by Vogelsang and others (Link* and Weissf think 
 otherwise), then we must regard the longulites as merely 
 cylindrical glass rods or spieulae, in which case the comparison 
 of a fibre of spun glass with a longulite becomes admissible. 
 If, however, we regard globulites merely as pellets of vitreous 
 matter it is difficult to see how glass can become devitrified 
 through their development, and, that these minute bodies are 
 endowed with properties which mere glass pellets do not possess, 
 seems probable from their methods of grouping, and from the 
 double-refraction resulting from their coalescence. 
 
 Spherulitic structures have been induced in glass, not merely 
 through the agency of dry heat, but also by heating under con- 
 siderable pressure and in presence of water. Such experiments 
 made by Prof. DaubreeJ approximate, perhaps, more closely to 
 natural conditions than the dry heat process, but the results in 
 both cases are interesting since they show that similar structures 
 may be produced under somewhat different conditions. 
 
 Spherulites, as a rule, show no distinct nucleus, the crystalli- 
 sation appearing to have started from a point, but in some cases a 
 crystal or a group of crystals may be seen to form the nucleus. 
 Slight differences in the chemical composition of artificial glass 
 have been found by Mr. Herman to influence the production of 
 spherules, some kinds de vitrify ing throughout without the for- 
 mation of a single spherule, while in others of only slightly 
 different composition spherules are developed in profusion. || 
 
 The globulitic portions of some of the spherules in the obsidians 
 from the Liparis, already mentioned, are of a deep brown colour 
 when viewed in thin section by ordinary transmitted light. If 
 this colour be inherent in the globulites themselves it would 
 indicate that either there is some difference chemically, or in 
 degree of saturation, between these bodies and the surrounding 
 glass, or that the molecular condition of the former is more dense 
 than that of the latter, in which case a greater number of 
 molecules would occur in the globulitic portions of the section 
 than in the glass which is comparatively free from them, the 
 result being equivalent to an augmentation of tint produced by 
 the superposition of similarly coloured plates. Rocks are some- 
 times partially devitrified by the development of globtilite-.l 
 
 Spherulites are occasionally seen to be surrounded by glassy 
 borders**" (Fig. 1, PI: 1). They more frequently, however, have 
 no such borders, and possess merely a radiating crystalline struc- 
 ture (Fig. 3<7 S PI. 1), which is best, and often only, revealed 
 
 * Fogg. Annal. XLVI. Ueber die erste Eutstehung cler Krystalle, 1839. 
 f Ib. CXLIL, p. 324. 
 
 j Etudes Synthetiques de Geologic Experimental, p. 169, et f>cq. 
 Quart. Journ. Geol. Soc., Vol. XLL, 1885; Presidential Address by Prof. 
 Bonney. 
 
 ; j Herman & Kutley, Proc. Koyal Soc., Vol. XXXIX, 1S85. 
 " Zirkt'l, Mik. Beschaff, d. Min. u. Gest, p. 274. 
 ** Rutley, Monthly Microscopical Journ., April 1870. 
 
8 SPHERULITIC STRUCT DKE. 
 
 in polarised light, and, like other radiating crystalline aggregates, 
 they show a dark interference-cross when viewed between crossed 
 nicols (Figs. 3b, and 9, PL 1). This is the most common type of 
 spherulite. 
 
 Less common, and usually of considerably larger size, are those 
 which show no indication of a radiate structure, but appear 
 under tolerably high powers to be made up of minute granules 
 (Fig. 5, PL 1). 
 
 Sometimes spherules exhibit irregular, or denticulate, or serrated 
 boundaries (Fig. 4, PI. 1), as in some of the devitrified lavas of 
 the Glyders in North Wales. These have always a radiating 
 crystalline structure, and it appears as though their peripheral 
 development had been influenced irregularly by the crystallisation 
 of the surrounding matter.* 
 
 Occasionally elongated spherules are met with, which Zirkel 
 has termed axiolites, (Fig. 13, PL 1). Axiolitic structure has not 
 hitherto been seen in British rocks. The largest spherulitic 
 bodies often present quite a nodular character. These are the 
 " lithophysen " of Von Richthofen,f who considered them to have 
 resulted from the inflation of siliceous segregations by imprisoned 
 steam. They are frequently cavernous. 
 
 Professor Bonney, who, among others, has investigated the 
 nature of these bodies, has arrived at the following conclusions : J 
 <e That the nodular or spheroidal structure has been produced in 
 two ways." 
 
 a. u By simple contraction and roughly concentric cracking of 
 
 the mass in cooling, being thus intermediate between the 
 perlitic structure common in glassy acid lavas and the 
 spheroidal structure common in basalt, which, so far as I 
 know, is rather rarer in the former." 
 
 b. " By similar contraction in cooling, which is determined by 
 
 the presence of a cavity, and produced as follows: 
 When the cavity is first formed we may regard the whole 
 viscid mass in the neighbourhood as in a state of equi- 
 librium between the various forces acting on the cooling 
 lava (contraction, &c.), and the pressure of the gaseous 
 contents of the cavity. As cooling proceeds (uniformly 
 suppose) the volume of the latter diminishes rapidly, and 
 its pressure against the walls of the cavity decreases. The 
 various forces are no longer in equilibrium, and the con- 
 tractile strain will be relieved by the formation of a crack, 
 roughly concentric with the cavity which, as we might 
 expect, is more regular than it in form." 
 
 " That the cavities are then filled, wholly or partially, by 
 infiltrated minerals in the usual way." 
 
 * Quart. Journ. Geol. Soc., Vol. XXXV., p. 508. 
 
 t " Jahrb. k. k., Geol. Reichsanst." I860. 
 
 J Quart. Journ. Geol. Soc., Vol. XXXVITL, p. 289. 
 
PERLITIC STRUCTURE, 9 
 
 " That the nodules, thus rendered more solid (and in other 
 cases from the effect of their form, aided perhaps by extremely 
 minute differences in texture, due to the disturbance of equilibrium 
 in cooling), produce the usual distortion of the cleavage planes 
 when the whole mass is compressed." 
 
 Some observers believe these bodies to be merely large spherulites. 
 Mr. Cole, 'in a recent paper,* adopts this latter view, regarding 
 them, at all events in some cases, as the result of devitrification 
 which took place during cooling, and he adduces among other 
 reasons, the presence of a radial structure and the occurrence of 
 smaller spherules within these large ones. He does not consider 
 that they have been formed subsequently to the consolidation of 
 the rock in which they occur. Analyses made by K. von Hauer 
 show that there is little difference between the chemical com- 
 positions of these bodies and that of the matrix in which they 
 occur. Specimens from the Glyders, Conway, and other localities 
 in North Wales are deposited in the Eock Collection of the Museum 
 of Practical Geology. Figures of the microscopic structure of 
 spherulitic lavas will be found in the following volumes of the 
 Quarterly Journal of the Geological Society, Vol. XXXVII., 
 PL XX., XXL; Vol. XXXIIL, PL XX.; Vol. XXXVIIL. 
 Plate X. ; Vol. XL., PL XVIII. ; and Vol. XLL, PL IV. 
 
 Perlitic Structure. The delicate spheroidal cracks often occur- 
 ring in vitreous and in devitrified lavas constitute what is known 
 as perlitic structure. These cradks are seldom or never quite 
 continuous so as to form perfect spheroids ; they are approximately 
 concentric, but overlap one another so as to produce a series of 
 somewhat irregularly concentric shells, rather like the layers of 
 an onion. The rocks in which this structure occurs are usually 
 traversed by minute reticulating, rectilinear fissures, between which 
 the spheroidal, perlitic bodies lie (Fig. 14, PL Ij. These bodies 
 occasionally show depolarisation, due to strain, but this phenomenon 
 is only evident in moderately thick sections. It serve?, however, 
 to confirm what was previously surmised, namely, that perlitic 
 structure is the result of contraction/}- The perlitic fissures are quite 
 independent of other structures in the rocks in which they occur, 
 and are commonly seen to traverse streams of microliths, &c., 
 constituting the so-called fluxion structure, and they were probably 
 formed at the close of, or subsequently to, the period of solidification 
 of the vitreous rock. It is a structure not peculiar to the highly 
 silicated vitreous rocks, but also occurs in tachylyte, and is met 
 with in various amorphous or colloid substances, as in Canada 
 balsam, varnish, &:c. Although the perlitic fissures are often 
 extremely delicate, devitrification seems to fail, as a rule, to 
 
 * Quart Journ. Geol. Soc., Vol. XLL, p. 163. 
 
 f Quart. Journ. Geol. Soc., Vol. XXXII., p. 140., Vol. XXXIIL, p. 452, acd 
 Monthly Microscop. Journ., Vol. XV., p. 180. 
 
10 
 
 PERLITIC LAVAS. 
 
 obliterate them, the perlitic structure in some of the oldest-known 
 lavas being as perfectly preserved, in spite of devitrification, as in 
 glassy volcanic rocks erupted in comparatively recent times. 
 
 Haying now briefly reviewed the leading structural characters 
 of felsitic lavas and their modern representatives, we may proceed 
 to the description of some of the principal examples which have 
 hitherto been met with in England and Wales. The geologist 
 cannot hope to restore clearly the physical geography of the*far 
 distant ages when these felsites were molten streams, but it has 
 long been known that lava-flows of considerable extent, and 
 sometimes of considerable thickness, were poured out in very early 
 times, arid it has since been ascertained that those of the class 
 which forms the subject of this Memoir, cooled down into 
 rocks often identical with the obsidians, perlites, and pitch stones of 
 much later date, as was first indicated, from microscopical observa- 
 tion by Mr. S. Allport* in 1 877. 
 
 The figures on PL I. are not drawn from nature. They are 
 diagrammatic, but sufficiently like the structures they are in- 
 tended to represent, and will probably serve their purpose better 
 than actual drawings from microscopic preparations would have 
 done. 
 
 The figures in the other plates are given in little more than 
 outline, in order to suit the method of reproduction employed. 
 
 The microscopic textures and small spherulitic structures are 
 not as a rule appreciable by ordinary transmitted light, while the 
 method of shading adopted being quite inadequate to express the 
 appearance of most of these structures in polarised light, it has 
 been thought better to refrain from any attempt to represent them 
 except in the diagrammatic sketches on PL I. 
 
 References are, however, given in the text to works in which 
 these phenomena are properly figured, and in addition to these, 
 admirable drawings will be found in Zirkel's " Microscopical 
 Petrography of the 40th Parallel," and Fouque and Levy's 
 " Mineralogie Micrographique." 
 
 In the following pages the term " perlite ' is not employed as a 
 rock-name, since perlitic structure may be developed in any 
 vitreous rock. 
 
 Many of the lavas here mentioned might be described as 
 Rhyolites, or lavas which contain from over 60 to 80 per cent, of 
 silica, and which consist chiefly of orthoclastic felspar (sanidine), 
 more or less quartz, occasionally tridymite, and usually some 
 vitreous matter, or they may be wholly vitreous. Such rocks 
 nearly always show fluxion structure. The term rhyolite includes 
 
 * Quart. Journ. Geol. Soc., Vol. XXXIII., p. 449. " On certain ancient 
 Devitrified Pitchstones and Perlites from the Lower Silurian District of Shropshire." 
 
SHROPSHIRE. 11 
 
 crystalline volcanic rocks, such as quartz-trachytes, and, in a 
 sub-group, the " hyaline rhyolites " or vitreous lavas of similar 
 chemical composition. None of those now occurring in England 
 and Wales retain their original glassy character. Since the 
 devitrification of hyaline rhyolites approximates to, or is identical 
 in character with the crystallization of those rhyolites which have 
 assumed a crystalline structure at the time of their solidification 
 from fusion, it is often extremely difficult, in the absence of perlitic 
 structure, to distinguish between the members of the two series. 
 In the following descriptions the name rhyolite, although a bad 
 one in some respects, will be employed in doubtful cases. 
 
 SHROPSHIRE, &c. 
 
 In localities where volcanic outbursts have occurred in recent 
 or comparatively late geological periods, as in the Liparis, and 
 the Yellowstone District, passages from crystalline to hyaline 
 rhyolites may be seen. 
 
 The presence of rocks of this class in England and Wales was 
 not recognised until Mr. S. Allport demonstrated their existence 
 in his admirable paper, already cited. Within a short time similar 
 rocks were recognised by other observers in the Silurian districts 
 of Wales, Westmorland, and Cumberland. They have since been 
 found in the Shetlands by Professor A. Geikie and Messrs. Peach 
 and Home,* in rocks of Old Red Sandstone age. Devitrified 
 rhyolites also occur in Jersey, and Professor Ch. de la Valle'e 
 Poussin has succeeded in correlating similar rocks occurring in 
 Belgium with those already met with in England and Wales.f 
 
 This eminent petrologist has arrived at the conclusion that the 
 rocks formerly described as eurites occurring at Grand-Maml, 
 near Gembloux, are certainly rhyolites, and he considers that 
 they have once been vitreous, in part. They are associated with 
 carbonaceous schists containing Climacograptus scalaris, while 
 other beds in the immediate vicinity are replete with fossils 
 characteristic of the Caradoc or Bala series. In Professor Poussin's 
 work numerous comparisons are made of these ancient Belgian 
 rhyolites with those of England and Wales occurring on the same 
 stratigraphical horizon, the microscopic characters of the former 
 having been studied side by side with some of the most typical 
 sections figured and described in the foregoing pages. 
 
 Among the oldest rhyolites occurring in this country may be 
 mentioned those from the neighbourhood of Wellington, in Shrop- 
 shire, the principal exposures being at Lea Rock and Lawrence 
 Hill, and also at Charlton Hill, and Caer Caradoc, near Church 
 Stretton. They are associated with quartzites which were 
 
 * Trans. E. Soc. Edin., Vol. XXXII., Pt. II., p. 359. 
 
 f "Les Anciennes Rhyolites, dites Eurites, de Grand-Maml." Bulletins de 
 1' Academic royale de Belgique 3 me serie. t. x. No. 8, 1885. 
 
12 LAKE DISTRICT. 
 
 mapped by the Geological Survey as Garadoc Sandstone. Mr. 
 Allport, however, in common with many other observers, believes 
 these beds to be of considerably earlier date. Fragments of 
 rocks which Professor Bonney regards as devitrified rhyolites 
 occur in certain breccias in Charnwood Forest.* The rhyolite 
 first recognised as such by the same author, which occurs in the 
 vicinity of Llyn Padarn, in North Wales, is also considered by 
 him as a lava of Pre Cambrian age.j Similar rocks have also been 
 described by Professor Bonney as occurring N. W. of Llanddeiniolen, 
 in Caernarvonshire, and at Brithdir, near Bangor, in an appendix 
 to a paper by Professor T. McK. Hughes, " On the Pre-Cambrian 
 Kocks of Bangor." J Mr. T, Davies, in an appendix to a paper by 
 Dr. Hicks, has also described an agglomerate occurring at Caer- 
 bwdy, on the road from St. Davids to Solva, which consists 
 largely of rhyolitic fragments. Professor Bonney, in an appendix 
 to a paper by Dr. Call away, records the occurrence of a trachyte 
 at Carreg-winllan, Pensarn, S.E. of Am wlch, in Anglesey. || As 
 the trachytes are closely related to the felsitic lavas this is worth 
 noting here. 
 
 Leaving, now, these rocks, of which the precise age is doubtful, 
 and of which notices will be found in the papers cited, we may 
 turn to the description of similar lavas that were unquestionably 
 erupted during the Silurian period. 
 
 LAKE DISTRICT. 
 
 At RED CRAG, 1^ mile N.E. of Stockdale, Westmorland, 
 there is a delicately banded rock which, under the microscope, 
 shows very distinct perlitic fissures over the entire section 
 (PI. II., Fig 2). A finer example of this structure is seldom to 
 be seen. The devitrification is spherulitic and crypto-crystalline, 
 the former condition appearing to predominate, but the spherules 
 are exceedingly minute and only to bs detected on attentive 
 examination. Some isotropic matter (? micro felsitic base) occurs 
 in the section. The rock is a devitrified perlitic obsidian. It 
 is of Lower Silurian age. The specimen here described was 
 collected by Mr. Aveline. 
 
 At the northern end of the LONG SLEDDALE Valley, in West- 
 morland, and at several other places along the outcrop of the 
 Coniston Limestone, a compact and often pinkish felsite occurs, 
 which frequently shows a very delicate banding. 
 
 A specimen collected by the author at TILL'S HOLE, at the 
 northern end of the Long bleddale Valley, shows, under the 
 microscope, a well-marked perlitic structure, and is completely 
 devitrified, chiefly by the development of minute spherules with 
 
 * Quart. Journ. Geol. Soc., Vol. XXXIV., p. 207. 
 t Ib. Vol. XXXV., p. 309. 
 I Ib. Vol. XXXIV., p. 137. 
 & Ib. Vol. XXXIV., p. 153. 
 || Ib., Vol. XXXVII., p. 230. 
 
LAKE DISTRICT. 13 
 
 a, radiating crystalline structure. That this was once a perlitic 
 obsidian, or perlite, as some might prefer to call it, there can be 
 no doubt. The section ia traversed by delicate bands (fluxion 
 structure) which are crossed by the perlitic cracks (PI. II., Fig. 1). 
 Spherulitic devitrification followed the development of the perlitic 
 structure. 
 
 The rock from which this section was cut is of a pale pinkish- 
 buff, or light yellowish-brown tint, traversed by delicate brown 
 bands, like fine lines drawn in sepia and intersected by little veins 
 of quartz.* 
 
 In the English Lake District other felsitic lavas of Lower 
 Silurian age have also been collected by Mr. W. Talbot Aveline 
 and Mr. Thos. Hart. Those collected by Mr. Aveline are chiefly 
 from the district to the north-west of Coniston Lake ; while 
 Mr. Hart's specimens were derived from the neighbourhood of 
 GRIZEDALE TARN. One of the latter, showing a well-marked 
 columnar jointing, came from a spot 200 yards east of the tarn. 
 The rock is of a dark ashen greenish-grey tint, the joint surfaces 
 being chocolate-brown. The specimen is a somewhat irregular 
 six-sided prism, terminated at one extremity by a flat joint plane 
 which cuts the axis of the prism at an angle of about 15. 
 
 In thin section, under the microscope, this rock shows well- 
 defined perlitic structure. It was, therefore, evidently vitreous. 
 The section is pervaded by a deep green tint, due to the presence 
 of some mineral of secondary origin, probably chlorite. The 
 occurrence of this rock, as a lava, is not indicated on the 
 Survey Map ; but the exposure, according to Mr. Hart, is not a 
 large one.f 
 
 About three-quarters of a mile N.N.E. of SEATHWAITE CHURCH, 
 IN THE DUD DON VALLEY, a rhyolitic rock has been mapped by 
 Mr. Aveline, in which, under the microscope, a mottled streaking 
 occurs. These markings are not very strong, but they clearly 
 show the character of the rock. In the same section some cracks, 
 much more strongly defined, traverse the direction of flow in 
 straight and reticulating lines. The rock is of a dark brownish- 
 green colour, with minute greyish flecks. It has a slightly greasy 
 lustre and a somewhat conchoidal fracture. 
 
 In the CoPPEK-MINE VALLEY, N. OF CONISTON, a sage- 
 green faintly-mottled felsitic rock occurs. It has a conchoidal 
 fracture, and shows under the microscope a few indications of 
 perlitic structure. The groundmass of this rock transmits very 
 little light between crossed nicols. Numerous microliths and a 
 few porphyritic crystals of felspar occur in the section. The 
 latter are very clear and glassy. This, in conjunction with the 
 uniform texture of the groundmass, the faint evidence of perlitic 
 structure, and the conchoidal fracture of the specimen, indicate 
 that it was once a vitreous rock. 
 
 * Quart. Journ. Geol. Soc., Vol. XL., PI. XVIII. 
 
 I The columnar structure of the rock was noted by the late Mr. Ward on his MS. 
 Maps. 
 
 u 18642. B 
 
LAKE DISTRICT. 
 
 A pale-drab felsite, very compact, and with an irregular or 
 splintery fracture, occurs on THE KNOTT, BROUGHTON Moon. 
 It has a felsitic groundmass of fine texture, in which coarser 
 Isolated grains and aggregates of grains occur, either forming 
 p?eudomorphs after minute crystals, or wavy strings which follow 
 a general direction. The banded structure is best seen by 
 ordinary transmitted light, and it is very suggestive of fluxion 
 structure. Some of the felsitic lavas constituting part of CON- 
 WAY MOUNTAIN, in North Wales, are very similar in microscopic 
 character. 
 
 On the WESTERN SIDE OF GREAT STICKLE, the bottom bed 
 of the series appears, from both field evidence and microscopic 
 characters, to be a felsitic lava. 
 
 A felsite, which may or may not be a rhyolite, occurs at COLT 
 CRAG, on the east side of Coniston Old Man. 
 
 The foregoing rocks will be described in the Memoirs of the 
 Geological Survey, Explanation of Quarter-Sheet 98 N.W., in 
 which certain rhy elite-tuffs are likewise noticed. These tuffs 
 occur at the following localities : 
 
 1 . AT APPLETREEWORTH ; in beds which, according to Mr. 
 
 Aveline, lie conformably beneath the Coniston Lime- 
 stone. This is an exceedingly good example of a 
 rhyolite-tufF or obsidian-tuff. 
 
 2. S. W. OF CONISTON ; in a bed lying conformably to and 
 
 directly beneath the Coniston Limestone, and resting un- 
 conformably on the older volcanic series, according to 
 Mr. Aveline. 
 
 3. At the WEST END OF THE YE WD ALE BRECCIA ; where it is 
 
 lost below the Coniston Limestone, as stated by Mr. 
 Aveline. The fragments in this tuff appear to be only 
 partially rhyolitic. 
 
 A microscopic section made from a specimen collected by Mr. 
 Aveline on the south side of the Beck, 300 yards N.W. OF SPA 
 WELL, SHAP WELLS, shows very delicate and exceedingly con- 
 voluted banding, while in some parts crystallisation along certain 
 lines has given rise to structures closely resembling the axiolites 
 described by Zirkel. The section is partly cryptocrystalline, 
 and contains quartz both in porphyritic crystals with irregular 
 contours, in perfectly rounded crystals, and in crystals in which 
 the angles are only slightly rounded. When magnified about 
 fifty diameters parts of the section exhibit a damascened appear- 
 ance like the mottle on a gun-barrel. There seems little doubt 
 that the rock from which this specimen was derived is a felsitic 
 lava. 
 
 Among the specimens collected by the late Rev. J. Clifton 
 Ward is a felsite which passes " through the trap of KNOTT END." 
 It consists of greyish-green and deep red or brownish-red bands 
 and spots. Under the microscope one set of bands is seen to be 
 partly devitrified by crystalline grains and partly by microliths 
 
NOKTII WALE*. 15 
 
 while the others are much more transparent in thin section, have 
 a deep yellow or orange colour and a more glassy appearance, 
 and transmit far less light between crossed nicols. They are, 
 however, devitrified to a considerable extent by microliths. These 
 latter bands show distinct evidence of spherulitic structure, and 
 may be regarded as more or less continuous belts of spherules. 
 The rock is probably a devitrified spherulitic obsidian or pitch- 
 stone. 
 
 NORTH WALES. 
 
 The most numerous, and among them some of the best, 
 examples of felsitic and devitrified lavas are met with in different 
 parts of Wales. 
 
 Among the most striking peculiarities which these rocks pre- 
 sent may be mentioned the occurrence in some of them of large 
 spherical or spheroidal nodules. Bodies of this kind have already 
 been alluded to on p. 8. 
 
 They are well developed in the felsite of ESGAIR-FELEN, a 
 spur of the GLYDERS, overlooking the PASS OF LLANBERIS. 
 
 A specimen of a similar rock was collected by the author some 
 years ago at BODLONDEB POINT, CONWAY. This is probably a 
 continuation of the nodular felsite mentioned by Prof. Bonney 
 as occurring in the DIGA>:WY HILLS on the opposite side of the 
 Conway Estuary, and described by him in his paper, " On 
 some Nodular Felsites in the Bala Group of'. North Wales."* 
 In the same paper other nodular fel sites have been noted between 
 CONWAY FALLS and PANDY MILL, in the LEDR VALLEY, about 
 half a mile above the new viaduct, where nodules from one to one and 
 a half inches, and sometimes nearly three inches in diameter occur, 
 and iii the upper part, and at the western end, of Conway Moun- 
 tain. In the last-named locality Professor Bonney states that he 
 believes the rock still retains a glassy residuum, a point of interest 
 in rocks of such great antiquity. 
 
 In addition to the spherulitic structures occurring in the lavas 
 of ESGAIR-FELEN a distinctly perlitic one has been described by the 
 author, thus leaving no doubt that the rock was once vitreous. 
 In one section from this locality the perlitic structure is very 
 curiously restricted to certain definite areas, Fig. 1, PI. IV., the 
 boundaries of which follow the contours of mosaic-like aggregates 
 of quartz, which may be regarded as filled-up vesicles. This 
 phenomenon is probably due to an originally different state of 
 tension in the glass immediately surrounding the vesicles. 
 
 Another felsitic lava, which crops out about 100 feet from the 
 base of the Giyder Fawr,f shows a small spherulitic structure under 
 
 * Quart. Journ. Geol. Soc., Vol. XXXVJIL, p. 289. 
 t /&., Vol. XXXV., p. 508. 
 
16 NORTH WALES. 
 
 the microscope,, the spherules, or rather radiating crystallisations 
 which appear to constitute the entire rock, having usually very 
 jagged boundaries (see Fig. 4, PI. II.). 
 
 In connexion \vith the larger spheroidal or nodular bodies it is 
 worth noting that a thin section through a concretion or spheroid, 
 about an inch in diameter, from one of the felsites of Skomer 
 Island, off the west coast of Pembrokeshire, is traversed by slightly- 
 flexed, approximately-parallel bands, some of them extremely 
 narrow and charged with nearly opaque ferruginous matter, while 
 other bands show a fine crypto-crystalline structure. The matter 
 lying between these bands is coarsely crystalline, this texture 
 contrasting strongly with that of the more delicate belts. These 
 spheroids, evidently of secondary origin, may be compared with 
 those occurring in the Magnesian Limestone of Durham, which, 
 in a similar manner, are traversed by the planes of lamination in 
 the limestone, while this Skomer Island concretion is traversed 
 by the bands common in viscous lavas. 
 
 Close to the GLYDERS and on the right-hand side of the road 
 ascending from PONT-Y-GROMLECH to GORPIIWYSFA a dark-grey 
 felsite occurs which has a fissile, platy structure, and a splintery 
 fracture. When examined microscopically it shows great numbers 
 of irregularly-shaped shreds and strings, (PI. II. Fig. 6). 
 
 In polarised light the rock is seen to be felsitic throughout, but 
 the crystalline texture of the shreds differs somewhat from that of 
 the groundmass of the section. A striking similarity is seen 
 between the microscopic structure of this rock and that of a 
 Hungarian Obsidian, from Tolcsva. A rhyolite from Gardiner's 
 River, Montana, U.S., also exhibits a similar structure.* It may 
 be worthy of note that the forms of these shreds, as seen in section, 
 bear a certain resemblance to the forms of shreds of glass met 
 with in some volcanic ejectamenta. There seems, however, but 
 little doubt that the Pont-y-Gromlech rock is a devitrified obsidian 
 or rhyolite, similar to the Tolcsva or to the Gardiner's River 
 rocks.f The structure is not by any means common, and appears 
 to be due to the extremely convoluted bands being cut through 
 in a given direction, so that only portions of the bands are in- 
 cluded between the planes of section. What appears to be a 
 similar structure, but on a much larger scale, occurring in a felsite 
 on the Minnesota shore, is figured by Mr. R. D. Irving in his 
 treatise " On the Copper-bearing Rocks of Lake Superior," where 
 he attributes it to tl flowing in a viscid condition.''^ 
 
 * Quart. Journ. Geol. Soc. Vol. XXX VII., p. 400. 
 
 f Professor Ch. de la Vallee Poussin figures and describes a precisely similar 
 rock, which occurs in beds approximately of Bala age at Grand -Manil, near Gem- 
 bloux, in Brabant. " Les Anciennes Khyolites, dites Euritcs, de Grand-Manil." 
 Bull. Acad. royalc de Bolg.. 3 mc serie, t. x., No. 8, 1885. 
 
 J Third Annual Kepoit, U.S. Gcol. Survey, p. 129. 
 
NORTH AVALE?. 17 
 
 A specimen of felsitic lava from CONWAY MOUNTAIN* shows a 
 well-marked banded structure. Under the microscope it presents 
 a very uniform felsitic appearance and shows some pseudomorphs 
 of exceedingly fine-grained felsitic matter, after felspar crystals. 
 Thess crystals or pseudomorphs lie with their longest axes parallel 
 with the banding of the rock (Fig. 5. : PI. III.). 
 
 At CLOGWYN DTJR AEDDU, a ridge about a mile to the N.W. 
 of the summit of Sxowpox, a well -banded devitrified obsidian 
 occurs (PI. IL, Fig. 5), showing faint, though unmistakeable traces 
 of perlitic structure, which on account of iis faintness renders this 
 section especially worthy of study.f It indicates, indeed, that 
 perlitic structure may become completely effaced by devitrification, 
 and that we may, therefore, often be justified in the assumption 
 that many felsites which now show no trace of perlitic structure 
 may once have been vitreous. Although we may look for this 
 structure as a trustworthy test of the former vitreous character of 
 a rock, yet we must not lose sight of the fact that there are vast 
 numbers of specimens of recent obsidians and other glassy rocks 
 which show no trace of perlitic structure, and it would therefore 
 be not merely injudicious, but erroneous, to exclude all felsites, 
 in which this structure is absent, from the category of vitreous 
 rocks. On this account some examples of what ara provisionally 
 termed " doubtful felsitic lavas " are appended to this monograph. 
 In connexion with this doubtful series it is interesting to note 
 the close resemblance between some of the claystone porphyries 
 (Thonsteinpo?phi/r) and certain devitrified rhyolitic rocks. Apart 
 from the devitrification which has taken place in the Clogwyn 
 dur Arddu lava, there is no structural distinction to be found 
 between it and ordinary banded obsidians. The structure of rocks 
 of this class may, however, be somewhat difficult to interpret at 
 times. In a section of a banded felsitic lava, similar to that from 
 Clogwyn dur Arddu, but derived from Vicart Point, on the 
 north coast of Jersey, the banding has been greatly disturbed by 
 crushing and displacement, no less than a dozen minute faults 
 occurring in one band within a space of less than an inch, while 
 fragments of gome bands abut against others at angles of about 
 45 degrees. The resemblance of the lavas of Clogwyn dur 
 Arddu to the obsidian of Ascension was long ago noted by Sir 
 Andrew Ramsay.J 
 
 On the south side of the Oapel Curig Road, about a quarter 
 of a mile from BEDDGELERT, Mr. G. G. Butler collected specimens 
 of a devitrified spherulitic obsidian or pitchstone. The rock is 
 dark greenish-grey, spotted with pale greenish-grey spherules, 
 some isolated and over J inch in diameter (PL II., Fig 3.), while 
 others have coalesced, forming bands ^ to J inch in breadth. 
 
 First quarry past the turnpike on the road to Penmaen Mawr. 
 f This structure is not indicated in Fig. 5, PI. II. It occurs in another part of 
 the section, and could only be properly rendered by a more delicate style of engraving. 
 J "Descriptive Catalogue of Rock Specimens," 3rd Edition, 18fi2,Vlus. Piact. GeoL 
 Quart. Journ. Geol. Soc., Vol. XXXVII., p. 404. 
 
18 SKOMER ISLAND. 
 
 Under the microscope the spherules are seen to have narrow 
 but clear and colourless, borders, very sharply defined, while 
 similar sharply-defined lines traverse the matrix in which the. 
 spherules lie. Some of these lines are straight, while others 
 describe curves which indicate a rude perlitic or spheroidal 
 structure. Perlitic markings on a smaller scale are also visible 
 in some parts of the section. Figures illustrative of these structures 
 are appended to the paper already cited. 
 
 On the eastern side of the EIFL RANGE, in Caernarvonshire, a 
 rhyolite occurs which has been described by Professor Bonney.* 
 
 SKOMER ISLAND. 
 
 The felsitic lavas of Skomer Island off the West Coast of 
 Pembrokeshire and south of St. Bride's Bay, must be regarded as 
 of Lower Silurian age, as indicated by Sir Roderick Murchison.f 
 Specimens from this locality, collected by Sir Andrew Ramsay, 
 are deposited in the Rock Collection in the Museum of Practical 
 Geology, and it is from sections taken from some of them that the 
 following notes have been made. 
 
 The felsitic lavas of this island are probably the (i quartzose 
 cornean, mostly striped," of Sir Henry Dela Beche,J who also 
 mentions the occurrence of a i( fine-grained compact greenstone 
 sometimes approaching to cornean." A. section made from one of 
 these (< greenstones " in the Jermyn Street collection shows that it 
 is a basalt. 
 
 These rocks are considered to belong to the Llandeilo or to the 
 Bala series. Their microscopic characters were first described in 
 1881. They show, as a rule, a well-marked banding ; while 
 occasionally they contain spherules an inch or more in diameter. 
 Under the microscope both spherulitic and perlitic structures are 
 beautifully defined. The microscopic spherules generally have a 
 radiating crystalline structure, but in the larger ones this is either 
 imperfectly developed, so that they show no dark cross between 
 crossed nicols, or else it is absent. These larger bodies sometimes 
 interrupt the bands composed of the smaller spherules. In the 
 section from which Fig. 1. PI. Ill,, was drawn, by ordinary trans- 
 mitted light, the finely-banded, perlitic and spherulitic structure of 
 the rock is admirably seen. The spherules appear colourless, but the 
 section is pervaded by clear brownish-green serpentine, seemingly 
 quite structureless, which lies between the perlitic cracks and com- 
 monly occupies the central area of each perlitic section, so that we 
 can only conclude that in most cases the nuclear portion of each 
 
 * Quart. Journ. Geol. Soc., Vol. XXXV., p. 305. 
 
 t " Siluria" 4th Edition, p. 144. 
 
 i Trans. Geol. Soc., 2nd Scries, Vol. II., p. 8. 
 
 Quart. Journ. Geol. Soc., Vol. XXXVII., p. 409. 
 
SKQMER ISLAND. 19 
 
 perlitic body was more readily soluble than the rest of the rock, 
 and that after its removal serpentinous matter refilled the cavity 
 and that similar infiltrations in percolating along the minute 
 planes of perlitic and laminar, or banded, structure formed 
 irregular deposits of serpentine, which substance represents, in- 
 deed, a large proportion of the rock in its present state. There is 
 not the slightest indication that this material has resulted from the 
 alteration of any magnesia-bearing mineral, so that we can only 
 assume that it has been brought in solution from some other 
 source. It may possibly be derived from basalt, which also occurs 
 in the island. The curious way in which this green matter 
 follows the minute structure of the rock, especially the perlitic, 
 picking it out in a manner which renders the structure unusually 
 striking, is a sufficient proof, if any be needed, of the secondary 
 origin of this substance. The rock is throughout highly spherulitic, 
 in fact spherules occur wherever serpentine does not occur. These 
 are generally very small, having a radiating crystalline structure 
 and showing dark crosses between crossed nicols. They appear 
 massed together in somewhat irregular belts of variable width, and 
 it is interesting to note that there is a very delicate banded 
 structure passing through them ; just such as one sees in modern 
 vitreous lavas, which is perfectly distinct, in spite of the 
 immense antiquity of the rock and the great changes which it has 
 undergone. 
 
 There are, in addition to the little spherules which constitute 
 so great a portion of the rock, some much larger ones. These 
 often interrupt the bands, composed of the little spherules, 
 quite abruptly. They show a tendency towards a radiate crystal- 
 line structure, but the crystallisation in these larger spherules is 
 very irregular, and they show no interference-cross between 
 crossed nicols. This is one of the most beautiful examples of a 
 devitrified lava to be met with in rocks of Lower Silurian age. 
 Another specimen from the same locality presents somewhat 
 different microscopic characters, no perlitic structure being visible ; 
 but, by ordinary transmitted light, the section is seen to" be 
 traversed by irregular bands, which look like segregations of brownish 
 or greenish-brown dust (Fig. 3, PI. III.), with numerous circular 
 or oval spots, which, under a power of about 250 diameters, 
 appear to be segregations of very fine chloritic scales. It is 
 evident, when the section is viewed between crossed nicols, that 
 these spots mark the position of spherules, which in many cases 
 have been obliterated ; and we may regard these dusty-looking 
 belts as spherulitic bands, while in some instances we may see 
 unaltered bands composed of spherules, the chlcritic matter having 
 segregated only along the margins of these bands. The broader 
 intermediate belts consist of crypto-erystalline matter with a few 
 included spherules and irregularly-shaped pale green or greenish- 
 brown plates of a chloritic mineral. 
 
 An additional example of what appears to be an altered spheru- 
 litic structure is afforded by the section shown in Fig. 2, PI. III. 
 
20 PEMBROKESHIRE. 
 
 Here, by ordinary transmitted light, we find the rock traversed by 
 approximately clear and colourless bands, irregularly moniliform, 
 often with adherent circular patches of similarly clear, colourless 
 matter ; while like circular areas, doubtless representing spherules, 
 lie scattered between these clear bands. The intercalated bands 
 consist of greenish-brown chloritic matter in intimate admixture 
 with fine doubly-refracting particle-. When the section is rotated 
 between crossed nicols it is evident that if the clear bands and 
 circular areas be sections through spherulitic belts and isolated 
 spherules, they have, at all events, parted with their radiating- 
 crystalline structure ; and in the case of certain narrow moniliform 
 belts, a strong jointed structure may be traced across them ap- 
 proximately at right-angles to the axis of the belt, the spaces 
 between each pair of joints being occupied by perfectly distinct 
 crystalline bundles, (Figs. 3 and 4, PI. IV.). This structure has 
 evidently been superinduced since the formation of the spherules, 
 and is precisely analogous to that met with in the devitrification 
 of thin, parallel-sided plates of artificially formed glass*. That 
 we have in this, and the preceding rock, examples of devitrified 
 spherulitic obsidians there can be little doubt. 
 
 PEMBROKESHIRE. 
 
 NEAR FISHGUARD and at STRUMBLE HEAD, on the north coast 
 of Pembrokeshire, certain felsitic rocks occur which have been 
 mapped by the Geological Survey as "altered Llandeilo flags." 
 Specimens so labelled, which have long been in the Rock Collection 
 in the Museum of Practical Geology, seemed to bear so close a 
 resemblance to devitrified lavas that sections of them were cut, and 
 subsequent microscopic examination indicates that the rock from 
 LLANWNDA, near Fishguard, is of a rhyolitic type. 
 
 It is crypto -crystal line, has a banded structure, and contains 4 
 small lath-shaped felspar crystals, ranged with their longest axes 
 approximately parallel to the banding (Fig. 6, PL III.). 
 
 In the spcimen from STRUMBLE HEAD, collected between the 
 Camp and the North Coast, the banding is absent, at all events in 
 the section examined, yet although there is no proof that it is a 
 rhyolitic rock, its microscopic character is not incompatible with 
 such a supposition, while its occurrence on approximately the 
 same horizon as the Llanwncla felsite lends support to the view 
 that it may be an extension of it. Under the microscope it is seen 
 to consist of a mesh-work or felting of delicate crystalline spiculae, 
 with apparently a little isotropic matter. It is a bluish-grey flint- 
 like rock or hornstone. A specimen, also in the Jerm^n Street 
 collection, labelled, " Altered Llandeilo Flags from Carneddu, near 
 Builth,'* presents a somewhat similar appearance under the micro- 
 
 <ji Figured in a paper " On the Microscopic Characters of some Specimens of 
 Devitrified Glass, -with notes on certain analogous Structures in Rocks." Hermann 
 and Rutley. Proc. Royal Soc., Vcl. XXXIX., p. 87. 
 
PEMBROKESHIRE. 21 
 
 scope, but, in this case, the section, when rotated between crossed 
 nicols, breaks up into a series of irregularly rounded and angular 
 patches, much more distinct than any in the Strumble Head 
 section. An examination of more specimens from this locality seems 
 needful before any decided opinion can be given concerning the 
 original nature of this rock. It is, indeed, only by carefully 
 studying the microscopic characters of a series of more or less 
 altered slates and flags, and by comparing them with felsites of 
 known origin and of different crystalline textures, that the diffi- 
 sulty of distinguishing the rocks of the one class from those of the 
 other can be fully realised. It would, in fact, seem that we have 
 passages from slates and flags into felsites, so that in the absence 
 of any structural peculiarities it is scarcely possible to distinguish 
 the one class of rocks from the other. 
 
 When we consider the nature of the minute particles of which 
 slates are composed, a point most ably demonstrated by Dr. Sorby,* 
 we need feel but little surprise at the occasional doubt which the 
 microscopist experiences when dealing with rocks of this kind, 
 while a comparison of hand specimens of felsitic lavas of the 
 hornstone type with altered slates and porcellanites of similar 
 aspect shows that, without the microscope, we may meet with 
 like embarrassment. 
 
 Mr. T. Davies, in describing specimens from the TREFFGARN 
 EOCKS, on the Fishguard Road, evinces great caution, stating that 
 " the macroscopical and microscopical characters of these rocks 
 are so remarkably like those of the halleflintas from Sweden, that 
 they are not to be differentiated by means of the microscope." 
 He is inclined to regard them as of sedimentary origin. t 
 
 Closely allied to these rocks are certain spherulitic quartz- 
 porphyries which, according to Professor A. Geikie, are probably 
 of Lower Silurian age. They occur in the ST. DAVIDS' district, 
 near NUN'S CHAPEL and CHURCH SCHOOL QUARRIES. They are 
 intrusive in slaty beds associated with granite, and their resemblance 
 to spherulitic rhyolites is very close, except in the coarser 
 crystallisation of the materials of the groundmass. These rocks 
 have been described by Professor A. Geikie in his paper, " On 
 the supposed Pre-Cambrian Rocks of St. Davids,'"J to which 
 microscopic drawings by Mr. F. W. Eudler (Figs. 7 and 8, 
 PL X.) are appended. Mr. T. Davies has also described these 
 and some other, felsitic rocks. It seems quite possible that these 
 spherulitic quartz-porphyries may represent the " feeders " from 
 which some of the Silurian rhyolites emanated. They certainly 
 appear to have a character intermediate between that of an 
 ordinary quartz-porphyry and a quartz-rhyolite. 
 
 The quartz-rhyolite of LLYX PADARS has already been alluded 
 to on page 12. It shows very distinct fluxion structure and 
 
 * Quart. Journ. Geol. Soc., Vol. XXXVI., Presidential Address, I860, 
 t Quart Journ. Geol. Soc., Vol. XXXV., p. 292. 
 I Quart. Journ. Geol. Soc., Vol. XXXIX., p. 315. 
 lb., Vol. XXXIV., p. 164, and Vol. XXXV., p. 293. 
 
22 POST-SILURIAN. 
 
 contains numerous crystalline grains of quartz, which sometimes 
 include small fluid lacunae. The rhyolitic character of this rock 
 was first recognised by Professor Bonney. 
 
 POST-SILURIAN FELSITIC LAVAS. 
 
 No felsitic lavas of Post- Silurian age have yet been recorded in 
 England or Wales, with the exception of one occurring at BRENT 
 TOR, in Devonshire,* described by the author some years ago as a 
 devitrified rhyolite or rhyolitic breccia showing delicate sinuous 
 bands which appear to indicate fluxion structure, but the rock, 
 like most of those in the vicinity, is much altered and infiltered 
 with silica, and the evidence of its rhyolitic character is obscure. 
 It contains numerous opaque scoriaceous lapilli, and is either of 
 late Devonian or early Carboniferous age, probably the former. 
 Several other banded felsitic rocks, also from Devon, which closely 
 simulate felsitic lavas, have been cut and examined microscopically, 
 but they have proved to be merely altered sedimentary deposits. 
 A rock from the Mendips, described long ago by the author as 
 a <( devitrified pitchstone porphyry,"f is simply a porphyrite with 
 some devitrified matter in the groundmass. In the SHETLANDS, 
 true rhyolites of Old Red Sandstone age have been met with by 
 Professor A. Geikie and Messrs. Peach and Horne.J 
 
 Felsitic lavas, both of the Old Eed Sandstone period and also of 
 Tertiary age, have been noted by Professor Judd as occurring in 
 the district of LORNE and in the ISLE of MULL. He states that in 
 their weathered condition they sometimes show a spherulitic or 
 perlitic structure. Furthermore, truly vitreous lavas of Tertiary 
 age occur in the island of EIGG, in ARDNAMURCHAN, &c., while 
 the well-known pitchstones of ARRAN are probably referable to 
 the same period. 
 
 Porphyrites, sometimes with a vitreous base, are met with in the 
 extreme North of England. As these cannot be classed with the 
 lavas which form the subject of this monograph, and since the 
 rhyolites of Old Eed Sandstone and later age lie outside the 
 boundaries of England and Wales, they can here claim merely a 
 passing notice. 
 
 FELSITES OP DOUBTFUL ORIGIN. 
 
 A felsite from the WEST OF LLYN ARENIG, six miles west of 
 Bala, appears from its microscopic characters to be a devitrified, 
 banded obsidian (PI. III., Fig. 4). It has a very uniform crypto- 
 crystalline structure. The bands are of finer texture than the 
 
 * Memoirs of the Geol. Survey of England and Wales, " The Eruptive Rocks of 
 Brent Tor," p. 32. PI. VII., Fig. 1. 
 
 f Mem. Geol. Survey, " Geology of E. Somerset and the Bristol Coal-fields "; 
 Appendix I., p. 208 ; PI. VII. 
 
 Trans. Royal Soc., Edin., Vol. XXXII., part 2. 
 
 Quart. Journ. Geol. Soc., Vol. XXX., p. 220. 
 
FEL5ITES OF DOUBTFUL ORIGIN. 23 
 
 rest of the rock. It contains a few small felspar crystals, some of 
 which are triclinic. No perlitic structure is visible, so that there 
 is no proof i\\&i it was ever a vitreous rock. 
 
 A felsite from MOEL-Y- MENTIS, two miles S.S.E. of Arenig, 
 shows no microscopic structures which permit the formation of a 
 definite conclusion as to its origin. The same might be said of a 
 felsite from the east side of Y. Graig, about eight miles S.W. of 
 Bala, except that in this case fragments of other rocks and broken 
 felspar crystals appear to be present, so that it is probably a 
 tuff. A section made from a specimen from the top beds on the 
 EAST SLOPE OF ARENIG, in Merionethshire, affords no satisfactory 
 microscopic evidence as to its origin. It is essentially a felsite, 
 and may have been either a very fine-grained volcanic ash or a 
 lava. By ordinary light it shows peculiar markings rather like 
 those in 'the Pont>-y-Gromlech rock (p. 16). Irregular spots of 
 green matter are present, which, from their form, may be pumice 
 fragments, filled with chlorite*, but there is no good proof of this. 
 No doubt an examination of more specimens from the same beds 
 might give a clue one way or the other. 
 
 The felsite of MYNTDD NODOL, 5 miles W.N.W. of Bala, is a 
 rock of much the same microscopic character as the preceding, 
 but has rather more the aspect of a lava. There seem to be 
 indications of fluxion structure, but they are very obscure. 
 
 A felsite which, under the microscope, shows a considerable 
 number of porphyritic quartz crystals around which sweep films 
 of a chloritic or sericitic mineral, occurs at Y-GARN, neat 
 Llanrhaidr-yn-Mochnant. It may be a quartz-rhyolite. 
 
 Between TAX- Y- GUIS IAN and CWM ORTHIN, near FFESTINIOG, 
 and at CAREGLWYD, LLANFECHELL, ANGLESEY, felsites of 
 doubtful origin, so far as microscopic evidence goes, also occur. 
 
 Many other felsites might be mentioned concerning which it is 
 difficult or impossible to give any decided opinion from their 
 microscopic characters. Some may be lavas, and some altered tuffs ; 
 while others may be porcellanites, or less metamorphosed slates 
 and shales, and, at times, even fine-grained felspathic grits which 
 have undergone more or less change may, from their mineral 
 constitution, be included in the somewhat elastic group of 
 felstone?. 
 
 A very good example of the obliteration of structural pecu- 
 liarities in rocks of this class is seen in a microscopic section of a 
 little felsitic vein, a friction-breccia about five inches broad, passing 
 through fine-grained Silurian grits in Smooth Beck, N.E. of 
 ESTHTVAITE WATER, in Westmorland, in which by ordinary trans- 
 mitted light the forms of angular and rounded fragments are 
 shown only by pale brown, nearly washed-out-looking patches, 
 which, between crossed nicols, are scarcely to be distinguished 
 from their felstic cement. Similar effacement of fragmental 
 
 * See description of pumice fragment by Sorby. Presidential Address, Quart. 
 Journ. Geol. Soc., Vol. XXXVI., p. 80. 1880. 
 
24 
 
 structure may also be noted in highly altered volcanic agglo- 
 merates. 
 
 So deceptive, sometimes, is the aspect of altered clastic rocks, 
 that it is scarcely possible to distinguish by the eye alone between 
 them and felsitic lavas, especially when the specimens possess 
 well-marked lamination and the surfaces have been somewhat 
 weathered. 
 
 The foregoing pages probably give only a few of the English 
 and Welsh localities where felsitic lavas occur, although it is 
 hoped that most of the spots where they are at present known 
 have here been recorded. No doubt, by following along the 
 strike of felsitic beds now known to be lavas, the list of localities 
 might be added to considerably ; while further examination of 
 felsites of doubtful origin in the field may lead to the recognition 
 of many old lava-flows which have hitherto passed unnoticed. 
 In doubtful cases the evidence gathered in the field is quite as 
 valuable as, or more so than, that derived from microscopic 
 investigation. 
 
25 
 
 TABULAR STATEMENT OF PETROLOGICAL AFFINITIES. 
 
 The subjoined table shows the relation which the rocks 
 mentioned in this Memoir bear to one another. 
 
 Crystalline. A 
 glassy resi- 
 duum, rare. 
 
 More or less 
 completely 
 vitreous ; but 
 crystallites, 
 as a rule, 
 abundant. 
 
 Felsitic Lavas = Devitrified (i.e. t crystalline) conditions of Obsidia or 
 Pitclistone. 
 
 Obsidian = The most perfectly 
 glassy condition 
 into which the 
 same moken matter 
 which produces 
 quartz - trachyte or 
 quartzless trachytes 
 can pass. Contains 
 little or no water. 
 Silica = 60 to 80 / . 
 
 Pitchstone 
 
 A less perfectly glassy 
 condition of tra- 
 chytic matter. Con- 
 tains a variable 
 amount of water up 
 to 10 %. 
 Silica = 63 to 73%. 
 
 Crystalline. 
 Commonly with 
 more or less 
 glassy matter. 
 
 Qua rtz- Trachyte. 
 Rhyolite of Richthofen. 
 Liparite of Roth. 
 Silica = 75 to 77 %. 
 
 i 
 Sanidine- Trachyte. 
 .....Sanidine-Rhyolite 
 
 Silica = 74 to 78 %. 
 .... Quartz-Pornhvru. 
 
 Trachyte. 
 ....Quartzless-Trachyte. 
 Uomite. 
 Silica = 62 to 68 %. 
 
 Crystalline. 
 
 Elvan. 
 ! 
 
 j 
 Granite. 
 
 The following brief statement of the lithological characters of the rocks mentioned 
 may help to render the preceding table more generally intelligible : 
 
 Felslte. Micro-crystalline, crypto-crystalline, micro-felsitic. Seldom any true 
 
 Obsidian. Glass containing crystallites and occasionally crystals. Banded, 
 perlitic, and spherulitic structures common. 
 
 Pitchstone. Glass with vast quantities of crystallites and often many porphyritic 
 crystals. Banded, perlitic, and spherulitic structures common. 
 
 Quartz-Trachyte > Rhyolite, or Liparite. Micro-crystalline or aphanitic ground- 
 mass, occasionally micro-granitic. Porphyritic crystals of quartz and sanidine, 
 also triclinic felspars, hornblende and biotite. 
 
 Sanidine- Trachyte, Sanidine-Rhyolite. Micro-crystalline or aphanitic ground- 
 mass, chiefly composed of felspars. Porphyritic felspars and biotite. Tridy- 
 mite often present. 
 
26 
 
 Trachyte, QuartzJess Trachyte, Domite. Crystalline groundinass generally, 
 consisting chiefly of felspar microliths with a very small quantity of glass. 
 Porphyritic sanidine, and often some tricliuic felspars, hornblende, and Hotite. 
 High per-centage of silica usually regarded as due to tridymite. 
 
 Quartz-Porphyry, Quarlz-Felsite, Elnan, Micro-crystalline (felsitic) matrix. 
 Porphyritic crystals of .quartz and orthoclase. Mica seldom present, and then 
 only in small quantity. 
 
 Granite. Crystalline (completely). Essentially composed of crystals of quartz, 
 felspar (chiefly orthoclase) and mica. 
 
 Obsidians and pitchstones occur both as lava flows and as dykes. 
 
 The trachytes occur chiefly as lava flows. 
 
 The quartz-porphyries form dykes. 
 
 Granite occurs in large intrusive masses and in veins. 
 
27 
 
 ALPHABETICAL LIST OF LOCALITIES. 
 
 Anglesey 
 
 Appletreeworth Valley - 
 Arenig, E. slope of 
 
 Beddgelert 
 
 Bodlondeb Point, Conway 
 Brithdir, near Bangor 
 Broughton Moor - 
 
 Caerbwdy, St. Davids 
 
 Caer Caraioc, near Church Stret- 
 ton 
 
 Careglwyd, Llanfechell, Anglesey 
 
 Careg-winllan, Anglesey - 
 
 Charlton Hill, near Church Stret- 
 ton 
 
 Charnwood Forest 
 
 Church'School Quarries, St. Davids 
 
 Church Stretton, Shropshire 
 
 Clogwyn dur Arddu, Snowdou - 
 
 Colt Crag, E. side of Coniston 
 Old Man 
 
 Coniston Lake, N.W. of - 
 
 Coniston Old Man 
 
 Conway Falls 
 
 Conway Mountain 
 
 Copper-mine, Valley N. of Conis- 
 ton - 
 
 Cwm Orthin, Ffestiniog - 
 
 Diganwy, near Conway - 
 Duddon Valley 
 
 Eifl Range 
 Esgair-felen, Glyders 
 
 Fishguard, Pembrokeshire 
 
 Glyders, Llanberis Pass - 
 Gorphwysfa, Llanberis Pass 
 Great Stickle 
 Grizedale Tarn - 
 
 Knott End 
 
 Kuott, The, Broughton Moor 
 
 
 Page. 
 12 
 
 Lawrence Hill, near Wellington 
 
 Page. 
 11 
 
 .. 
 
 14 
 
 Lea Rock, near Wellington 
 
 11 
 
 _ 
 
 23 
 
 Llanberis Pass - 
 
 8, 15 
 
 
 
 Llanddeiniolen 
 
 12 
 
 
 1 *7 
 
 Llanwnda, near Fishguard 
 
 20 
 
 ; 
 
 1 / 
 
 15 
 12 
 14 
 
 Llyn Arenig - - 
 Llyn Padaru 
 Long Sleddale, Westmorland 
 
 22 
 12,21 
 12 
 
 
 12 
 
 Moel-y-menyn - 
 
 23 
 
 tret- 
 
 
 MvnyddNoclol - 
 
 23 
 
 
 11 
 
 Nun's Chapel, St. Davids 
 
 21 
 
 lesey 
 
 23 
 
 
 
 _ 
 
 12 
 
 
 
 >tret- 
 
 
 Old Man, Coniston 
 
 14 
 
 _ 
 
 11 
 
 
 
 _ 
 
 12 
 
 
 
 Davids 
 
 21 
 11 
 
 Pandy Mill, near Conway 
 P on t-y- Cromlech, Llanberis Pass 
 
 15 
 16 
 
 Oil - 
 
 17 
 
 
 
 iston 
 
 14 
 
 Red Crag, near Stockdale, West- 
 
 
 - 
 
 13 
 
 morland 
 
 12 
 
 _ 
 
 14 
 
 
 
 _ 
 
 15 
 
 
 
 - 
 
 17 
 
 St. Davids, Church School Quar- 
 
 
 onis- 
 
 
 ries - 
 
 21 
 
 _ 
 
 13 
 23 
 
 St. Davids, Xuii's Chapel 
 Seathvraite Church, Duddon Valley 
 
 21 
 13 
 
 
 
 Shap Wells 
 
 14 
 
 
 15 
 
 Skomer Island - 
 
 16,18 
 
 
 13 
 
 Snowdon 
 
 17 
 
 
 
 Stockdale 
 
 12 
 
 
 
 Strumble Head - 
 
 20 
 
 _ 
 
 18 
 
 
 
 _ 
 
 15 
 
 
 
 
 
 Till's Hole, LongSleddale 
 
 12 
 
 - 
 
 20 
 
 Treffgarn Rocks - 
 
 21 
 
 - 
 
 8, 15 
 16 
 
 Wellington, Shropshire - 
 
 11 
 
 . 
 
 14 
 
 
 
 - 
 
 13 
 
 Y-Garn - 
 
 23 
 
 
 
 Y-Graig, S.W. of Eala - 
 
 23 
 
 _ 
 
 14 
 
 
 
 r 
 
 14 
 
 
 
EXPLANATION OF THE PLATES. 
 
 PLATE I. 
 
 Illustrative of structures occurring in vitreous and devitrified rocks. 
 
 The figures in this plale are diagrammatic, and represent thin sections, as seen 
 under various magnifying powers, ranging from 10 to 50 linear. 
 
 Fig. 1. Spherule, "with radiating crystalline structure and vitreous border. 
 
 2. Spherule, with similar structure to No. 1, but with a broad border formed 
 of globulites. 
 
 3. Spherules, with radiating crystalline structure, but without borders 
 
 a. As seen by ordinary transmitted light. 
 
 b. As seen in polarised light between crossed nicols-prisms. 
 
 4. Spherulitic body, with radiating crystalline structure and irregularly in- 
 dented surface. Not a spherule in form, but with the same radiating 
 crystalline structure common in spherules. 
 
 5. Spherule, with granular structure. 
 
 C. Globulites segregating, so as to form what appears to be an incipient 
 spherule. 
 
 7. Bands composed of mieroliths, deflected by a larger crystal (fluxion 
 structure). 
 
 8. Bands, showing different degrees of fineness in their crystalline structure. 
 Such bands may be micro or crypto-crystalline. 
 
 9. Bands composed of small spherules, as seen between crossed nicols. 
 
 10. Intricately convoluted bands (damascene structure). 
 
 11. Moniliform band, formed by the coalescence of spherules. 
 
 12. Parallel-sided band, cylindrical or tabular, formed by the coalescence of 
 spherules. 
 
 13. Axiolites. 
 
 14. Perlitic structure. 
 
 PLATE II. 
 
 1. Devitrified perlitic obsidian. Till's Hole, N. end of Long Sleddale 
 Valley, Westmorland. Devitrified by spherules, which are only appre- 
 ciable in polarised light. bb. Banding traversed by perlitic cracks. 
 gq. Quartz-veins, x 32. 
 
 2. Devitrified perlitic obsidian. Ked Crag, 1^ mile N.E. of Stockdale, 
 Westmorland. Shows perlitic structure, and is devitrified partially by 
 spherules, x 18. 
 
 3. Devitrified spherulitic pitchstone or obsidian. Beddgelert, N. Wales. 
 Coarse spherulitic structure. Perlitic structure present, but not shown 
 in drawing, x 18. 
 
 4. Spherulitic lava. Glyder Fawr, N. Wales. Made up of irregularly shaped 
 spherulitic bodies, x 77. 
 
 5. Devitrified obsidian. Clogwyn di'ir Arddu, Snowdon. Roughly parallel 
 banding. Very faint perlitic structure present, but not "shown in 
 drawing. x 20. 
 
 ,, 6. Devitrified obsidian or rhyoliie. Between Pont y- Cromlech and Gorph- 
 wysfa, Pass of Llanberis. Sections through convoluted bands, x 77. 
 
29 
 
 PLATE III. 
 
 Fig. 1, 2, and 3. Devitrified perlitic and spherulitic obsidians or pitchstones. 
 Skomer Island, off the West Coast of Pembrokeshire. 1 and 2, x 18 ; 
 3, x55. 
 
 4. Rhyolite or devitrified obsidian ? W. of Llyn Arenig, 6 miles W. of Bala. 
 
 x!8. 
 5. Rhyolite or devitrified obsidian. Conway Mountain, N. Wales, x 18. 
 
 6. Devitrified obsidian or rhyolite. Near Llanwnda, Fishguard, Pembroke- 
 shire, x 18. 
 
 PLATE IV. 
 
 1. Devitrified perlitic obsidian. Esgair-felen, Y. Glyder Fawr, North Wales, 
 x 10. 
 
 FF = Felsitic matter like PP between crossed nicols. 
 PP = Areas in which perlitic structure occurs. 
 qq = Aggregates of quartz. 
 
 2. String of spherules in devitrified obsidian. Skomer Island. Ordinary 
 iilumination. 
 
 3. The same in polarised light, showing transverse joints, separating distinct 
 crystalline bundles. 
 
 LI 18642. 
 
30 
 
 PLATE I. 
 
 tl 18642, 
 
31 
 PLATE II. 
 
 Fig. 1 
 
 Fig. 5. 
 
 Fig. 6. 
 
32 
 
 PLATE III. 
 
 Fig. 2. 
 
 Fig. 3. 
 
 Fig. 4. 
 
 Fig. 5. 
 
33 
 
 PLATE IV. 
 
 . fl. 
 
L O N D O*N : Printed by EYRE and S P o i T i s w o o r> F., 
 Printers to the Queen's most Excellent Majesty. 
 For Her Majesty's Stationery Office, 
 [8917. 375. 12/85.] 
 

 14 DAY USE 
 
 RETURN TO DESK FROM WHICH BORROWED 
 EARTH SCIENCES LIBKAKY 
 
 This book is due on the last date stamped below, or 
 
 on the date to which renewed. 
 Renewed books are subject to immediate recall. 
 
 LD 21-40m-10,'65 
 (F7763slO)476 
 
 General Library 
 
 University of California 
 
 Berkeley 
 
015 
 
 
98 NE - 
 96 NAV, 
 98 SE - 
 98 NE - 
 101 SE 
 104 SW, 
 108 SE 
 
 SHEET MEMOIRS OP THE GEOX.OGXCAX, SURVEY continued. 
 
 p<nm AT F HOSED \LE &e. Bv C. FOX-STRAITGWATS, C. REID, and G. BARKOW. is. Gd. 
 * m\3n$fr T ?TOtfand TH IRSK By C. FOX-STKAX &WATS, A. G. CAMEROX, and G. BARROW. 
 SW NORTHALLERT .\V-md KENDAL By W. T. AVEIINE, T. Me K. HUGHES, and R. H. TIDDEMAN. 
 
 ' K - H 
 
 (Note. .,y 0. T. 
 
 2s. 
 
 THF MINERAL DISTRICTS OF ENGLAND AND WALES ARE ILLUSTRATED 
 
 JYTH^FOLLOWINC PUBLISHED MAPS OF THE GEOLOGICAL su RVEY. 
 
 COAL-FIELDS OF ENSLAND AND WALES, 
 
 Durham continued. 
 
 Sheet Sheet. 
 
 Scale, one inch to a mile. 
 Anglesey, 78 (SW). 
 Bristol and Somerset, 19, 85. 
 Coalbrook Dale, 61 (NE & SE). 
 
 81 (NE),87 (NE, SE), 88 (SE). 
 
 9 25. Wolsingham. 
 10.' Edraondbycrs. -' Ki'ancepeth. 
 11. Ebchester. 30. Benny Seat. 
 !' Tantoby. 32. White Kirkley. 
 13. CHester-le-Street. 33. Hamsterley. 
 16. Hunstanworth. SJi. Whitworth. 
 17 Waskeriev. 38. Maize Beck. 
 18. Muggleswick. 41. Cockfield. 
 19. Lanchester. 42. Bishop Auckland. 
 
 Forest of ^^ n > **. /CT?) 55 (NE) 
 
 20. Hetton-le-Hole. 46. Ha^rksley Hill House. 
 
 . cm 6 / w\ 81 (NW) 89* 88 (SW, NW). 
 
 22. Wear Head. 52. Barnard Castle. 
 
 [jancasnirc, oil vii . ~ct^.* \^, /ivr\x7 p \* 
 
 23. Eastiratc. 53. Winston. 
 
 BB&nS^(^!^*3ifflSBiSR: 
 
 24. Stanhope. 
 Northumberland, 
 
 S. Staffordshire, 54 (N\V ^'^^U r \ 
 
 4k Rothburv. 80. Cramlington. 98. Walker. 
 
 S'lrcwsVjury, 60 (^*-E) 61 'JN \Y s o w / -__* 
 
 45 Lonsrframlinc- 81. Earsdon. 101. Whitfield. 
 
 ?VarwickslSre?62 (NESE),63(NW SW) , 54 (NE), R8 (NW). 
 
 Yorkshire, 88 (NE, SE), 87 (SW),92 (SE),93 (SW). 
 
 ton. 82. NE.of Gilsland. 102. Allendale 
 46. Broomhill. 83. Coadley Gate. Town. 
 47. Coquet Island. 87. Heddon. 103. Slaley. 
 
 
 54. Longhorsley. 88. Long Benton. 105. Newlands. 
 
 GEOLOGICAL MAPS. 
 
 55. Ulgham. ' 89. Tynemoutb. IOC. Blackpool Br. 
 56 Druridge Bay. 91. Greenheact. 107. Allendale. 
 
 Scale, six inches to a mile. 
 The Coal-fields and other mineral districts of the N. of 
 Eneland are published on a scale of six inches to a mile, 
 at S to 67 each. MS. Coloured Copies of other six-inch 
 mapVnoHntendedfo^ 
 nce in the Geological Survey Office, 2b, Jeimyn Street, 
 
 63. Nether \vittoh. 92. Haltwhistie. 308. Blanchland. 
 64. Morpeth. 93. HaydonB ridge. 109. Shotleyfleld. 
 65. NewbiBjgin. 94. Hexlio.m. 110. Wellhope. 
 72. Bedlingtou. 95. Corbridge. 111. Allonheads. 
 73. Blyth. 96. Horsley. 112. 
 97. Newcastle. 
 
 London. 
 
 Cumberland. 
 
 
 55. Searness. 69. Buttermere. 
 
 Lancashire. 
 
 56. Skiddaw. 70. Grange. 
 
 
 63. Thackthwaite. 71. Helvellyn. 
 
 S 15? Ireleth. ^ *Ormsl:irk, St. Johns, &c. 
 Tllvervtone. 85. Standish,&c. 
 
 64. Keswick. 74. Wastwater. 
 05. Dockrayc. 75. Stonethwaite Eell. 
 
 I!'. CartmeT 86. Adlin.ton Horwick, Ac. 
 
 "Westmorland. 
 
 22. Aldingham. 87. Boltoi ^e-Moora. 
 47. Clitheroe. *8. Bury 1 rey wood. 
 48 Colne, Twiston Moor. 89. Rpchdaie, &c. 
 
 2. Tees Head. 12. Patterdale. 25. Grasmere. 
 6. Dufton Tell. 18. Near Grasmere. 88. Kendal. 
 
 49. Laneshaw Bridge. 92. Bickerstaffe. 
 65. Whalley. 93. Wigan, Up Holland, &c. 
 66* Hawcate. 94. WestHoughton.Hindley. 
 67 Winewall. 95. Radcliffe, Peel Swinton. 
 61* Preston 96. Middleton, Prestwich. 
 62! Balderstone, &c. 97. Oldham, &c. 
 6S. Accrington. 100. Knowsley, Rainford, &c. 
 64. Burnlev. 10L Biltinge, Ashton, &c. 
 65. Stiperden Moor. 102. Lei-" 1 .*, Lowton. 
 69. Layland. 103. Ashley, Eccles. 
 70 Blackburn, &c. 10-*- Manchester, Salford, &c. 
 71*. Haslingden. 105. Asliton-uiider-Lyne. 
 *2. Cliviger, Bacup, &c. 106. Liverpool, &c. 
 73. Todmorden. 107. Prescott, Huyton, &c. 
 77. Chorley. 108. St. Helen's.BurtonWood. 
 
 Yorkshire. 
 
 116. Conistone 260. Honley. 
 7. Redcar. Moor. 261. Kirkburton. 
 9 133 Kirkbv i)2 Darton. 
 
 12*. Bowes. Malham. 203. Hemsworth. 
 13. Wvcliffe. 184. Dale End. 204. CampsalL 
 20. I ivthe. 185. Kildwick. 272. Holmfirth. 
 24. Kirkbv Ravens- 200. Keighley. 273. Penistone. 
 worth. 201. Bingley. 274. Barnsley. 
 5. Aldborough. 202. Calverley. 275. Darfield. 
 32. Whitby. " 203. Seacroft. 276. Brodsworth. 
 33. 204. Aberford. 281. Langsell. 
 33. Mr.rske. 215. Peeke Well. 282. Wortley. 
 39. Richmond. 216. Bradford. 283. Wath upon 
 
 78. Bolton-le-Moors. 109. Winwick, &c. 
 79. Entwistle. 111. Cheedale, Stockport, &c. 
 80. Tottington. 112. Stockport, &c. 
 81. Wardle. 113. Part of Liverpool, &c. 
 
 47.' Robin Hood's 218. Leeds. 284. Conisborough. 
 B-vv. 219. Kippax. 287. Low Bradford. 
 53. Downholme. 231. Halifax. 288. Ecclesfield. 
 68. Levbourne. 232. Birstal. 289. Rotherham. 
 
 
 82. Kidstones. 233. EastArdsley. 290. Braithwell. 
 
 
 84 E. Witton." 234. Castleford. 293. Hallam Moors. 
 
 Durham. 
 
 97. Foxup. 246. Huddersfield. 295. Handsworth. 
 
 1. Rvton. 5. Greenside. 
 
 9S. Kirk Gill. 247. Dewsbury. 296. Laughton - en 
 
 2. Gateshead. 6. Winlaton. 
 
 99. Haden Carr. 248. Wakeiiuld. le-Morthen. 
 
 3. Jarrow. 7. Washington. 
 
 100. Lofthouse. 249. Pontefract. 299. 
 
 4. S. Shields. 8. Sunderland 
 
 115. Arncliffe. 250. Darrington. 300. Harthill. 
 
 MINERAL STATISTICS. 
 
 Embracing the produce of Coals, Metallic Ores, and other Minerals. By R. HUNT. From 1853 to 1857, inclusive 
 Is. 6d. each. 1^8, Part L, Is. Qd.; Part IL, 5s. 1859, Is. 6c/. 1860, 3s. Qd. 1861, 2s.; and Appendix, Is. 1862, 2s. 6d 
 1S63. 2.9. Qd. 1864,2s. 1865, 2s. 6d. 1866 to 1881, 2s. each. 
 
 (These Statistics are now published by the Home Office, as parts of the Reports oftht Inspectors of Mines.) 
 
 THE IRON ORES OF GREAT BRITAIN. 
 
 Parti. The North and North Midland Counties of England (Ont of print). Part II. South Staffordshire. Price Is, 
 Part 111, South Wales. Price Is. 3d. Part IV. The Shropshire Coal-field and North Staffordshire. 1*. 3d.