PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND FOR 1923. VOL. XXXV. ANTHONY JAMES CUMMING, Government Printer. Brisbane. 1924. Price: Ten Shillings. PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND FOR 1923. VOL. XXXV. ISSUED 29th FEBRUARY. 1924. The Authors of Papers are alone responsible for the statements made and the opinions expressed therein. Printed for the Society by ANTHONY JAMES CUMMING, Government Printer, Brisbane.. 1924. R-S A. Price: Ten Shillings. The Royal Society of Queensland. © Patron : HIS EXCELLENCY SIR MATTHEW NATHAN, P.C., G.C.M.G. OFFICERS, 1924. President : Dr. E. 0. MARKS, M.D., B.A., B.E. Vice-Presidents : Professor H. J. PRIESTLEY, M.A. ( ex officio). W. H. BRYAN, M.Sc. Hon. Treasurer : Hon. Secretary : E. W. BICK. W. D. FRANCIS. Hon. Librarian : Professor E. J. GODDARD, B.A., D.Sc. Hon. Editor : H. A. LONGMAN, F.L.S. Members of Council : E. H. GURNEY. Professor H. C. RICHARDS. E. H. SWAIN. R. A. WEARNE, B.A. C. T. WHITE, F.L.S. Trustees : R. H. ROE, M.A. Hon. A. J. THYNNE. Hon. W. F. TAYLOR. Hon. Auditor : Professor H. J. PRIESTLEY, M.A. ‘odtoD^Z CONTENTS. VOLUME XXXV. Page. No. 1. — Presidential Address. On the Development of Scientific Thought. B y Professor H. J . Priestley , M.A. Issued 27th June, 1923. . . . . . . 1 No. 2.— On a Tertiary Fossil Insect Wing from Queensland (Homoptera Fulgoroidea), with Description of a New Genus and Species. By R. J. Tilly ard, M.A., Sc.D. {Cantab.), D.Sc. {Sydney), C.M.Z.S., F.L.S., F.E.S. Issued 27th June, 1923 16 No. 3. — Observations Regarding the Life-Cycle of Certain Australian Blowflies. By Professor T . Harvey Johnston, M.A., D.Sc., and G. H. Hardy. Issued 27th June, 1923 21 No. 4. — New Cactus Bugs of the Genus Chelinidea (Hemip- tera). By John C. Hamlin, M.Sc. Issued 1st August, 1923 43 No. 5. — Some Further Observations on the Dawson River Barramundi (Scleropages leichhardtii) . By Thos. L. Bancroft, M.B. Issued 1st August, 1923 . . 46 No. 6. — An Unusual Tourmaline -Albite Rock from Enog- gera, Queensland. By W. H. Bryan, M.Sc. Issued 28th August, 1923 .. .. .. .. .. 48 No. 7. — Notes on the Essential Oils of Daphnandra aro- matica. By T. G. H. Jones, B.Sc., A.I.C., and Frank Smith, B.Sc., F.I.C. Issued 21st December, 1 923 61 No. 8.- — Contributions to the Queensland Flora. By C. T. White, F.L.S., and W. D. Francis. Issued 21st December, 1923 . . . . . . . . . . 63 No. 9. — Notes on the Physiography of Eastern New Guinea and Surrounding Island Groups. By C. H. Massey. Issued 21st December, 1923 . . . . 85 No. 10. — Permo-Carboniferous Volcanic Activity in South- ern Queensland. By Professor H. C. Richards, D.Sc., and W. H. Bryan, M.Sc. Issued 14th December, 1923 . . . . . . . . . . 109 No. Ill — The Queensland Inocerami Collected by M. Lum- holz in 1881. By F. W. Whitehouse, B.Sc. Issued 29th February, 1924 .. No. 12. — The Composition of, the Volatile Oil of the Leaf of Daphnandra aromatica Bailey. By T. G. H. Jones, B.Sc., A.I.C., and Frank Smith, B.Sc., F.I.C. Issued 29th February, 1924 Abstract of Proceedings List of Library Exchanges-.. ’ List of Members . . . 127 133 xviii. xxii. . rr: 'V • ' " . ■ S ■■ Proceedings of the Royal Society of Queensland for 1923. Presidential Address. By H. J. Priestley, M.A. ( Delivered before the Royal Society of Queensland , 26th March, 1923.) Before proceeding with the main topic of my address, I should like to refer to three matters mentioned in the Annual Report. You will have noticed how large a proportion of onr annual expenditure is devoted to printing. Like all other scientific societies we are badly hit by the increased cost of publication, and I would urge on members the need for continued efforts to maintain and increase our resources. A gratifying feature of last year ’s. work is the co-opera- tion between our Society and the Royal Geographical Society. It is to be hoped that we may see a gradual drawing together for mutual support of all kindred societies in Brisbane. In such a way we might diminish the evils of isolation, serious everywhere, but especially serious in a city far from the main centres of intellectual activity. During the past year we have lost by death two ordinary members and one corresponding member of the Society — Dr. Alfred Sutton, Mr. James Johnston, and Professor J. A. Pollock. At the outbreak of war in 1914 Dr. Sutton was principal medical officer of the Queensland Military District. He left Australia with the first expeditionary force as Lieutenant-Colonel in command of the Third Field Ambulance, and was present at the landing on Gallipoli. For his services throughout the war he was appointed a Companion of the Orders of the Bath and St. Michael and St. George, and a Knight of Grace of the Order of St. John of Jerusalem. Mr. James Johnston entered the service of the Depart- ment of Public Instruction in 1881 as a pupil teacher at Warner. He served as assistant teacher at Warner and R.S. B. 2 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Fortitude Y alley, and head teacher at. Aramac, Emu Yale, Ashgrove, Kangaroo Point, Bundaberg, and Townsville. He became a district inspector in January, 1914, and continued bis [work in that capacity till his death iast September. Professor Pollock was appointed to the Chair of Physics in the University of Sydney in 1899. He did not confine his energies to his own department or his own University, but was always willing to bear his share of the general scientific work of Australia. He was a Fellow of the Royal Society of London, and an ex-President of the Royal Society of New South Wales and of Section A of the Australasian Association. Many of ns who worked with him at the meetings of the Association will always remember with gratitude the encouragement which he gave to those younger and less experienced than himself. During the rvar Professor Pollock did valuable Work at the Front in connection with practical applications of the theory of sound. ON THE DEVELOPMENT OF SCIENTIFIC THOUGHT. During a visit to England some three years ago it was my privilege to dine with one of the most distinguished living mathematicians. In the course of conversation we discussed the published proceedings of a celebrated mathe- matical society, and my host said, i:What I object to in that publication is that only once in six months does it contain a paper that a man can understand.” This remark emphasizes a great difficulty and a great danger in modern scientific work. Recent advances have been so varied and so rapid that no man can keep in touch with all develop- ments, even in one branch of science alone. We are compelled to specialize and our sphere of work is confined to a comparatively small region of one of the main subdivi- sions of science. It is sometimes advisable,, however, to turn from the cultivation of our own small fields and look out over the landscape as a whole. Accordingly I propose to-night to direct your attention to some points in the past history of scientific thought and to a danger which appears to underlie some of the modern work. It is frequently asserted that Greek Science was based on metaphysical calculations unchecked by experiment or PRESIDENTIAL ADDRESS PRIESTLEY. 3 observation. Like most other generalizations this embodies an element of truth, but, if interpreted too literally, it is apt to be misleading. Burnet, in his history of Early Greek Philosophy, accounts for the origin of the idea in the fact that the records we possess are mainly statements of results. These by themselves, without any indication of the methods by which they were obtained, certainly convey the impression that they were dogmatically asserted opinions. We know, however, that the Greeks had at their disposal a large body of observational facts obtained by the Chaldiean astronomers and the Egyptian surveyors. Furthermore, there are indications that opinions that at first sight appear to be fallacious metaphysical assump- tions are, in actual fact, erroneous deductions from accurate observation. For example, Xenophanes asserts that “all things are earth and water.” We might be tempted to take this as evidence of the Greek tendency to formulate a system of Natural Philosophy based on arbi- trary assumption, were it not for the fact that we have other records in which it is stated that “Xenophanes said that a mixture of the earth with the sea is taking place and that it is gradually being dissolved by the moisture. Pie says that he has the following proof of this : — Shells are found in midland districts and on hills ; and he says that in the quarries at Syracuse has heen found the imprint of a fish and of seaweed, at Paros the form of an anchovy in the depth of the stone, and at Malta flat impressions of all marine animals. 7 ’ From this it appears that his assertion was not based on mere assumption but was a faulty deduction from observation. As another example, consider a view of eclipses that was current in the early days of Greek Astronomy : The sun is a bowl full of burning material which normally has its concave side towards us, but which is occasionally reversed. This is not a wholly unreasonable interpretation of observed phenomena in the infancy of scientific thought ; the prominences visible during a solar eclipse might quite well be taken for flames showing over the edge of the inverted bowl. It is certain, then, that in some cases Greek scientific views arose as interpretations of experimental evidence; 4 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. interpretations, it is true, which appear fanciful and extrava- gant in the light of modern knowledge but which were not unreasonable at the time they were put forward. It may quite well be argued that Greek scientific method was not, as is commonly believed, pure metaphysical speculation, but rather the modern method in its early stages of development. The formulation of hypotheses to explain observations, one of the main features of modern work, was certainly practised by the Greeks and probably originated with them; for there is nothing to suggest that Babylonian or Egyptian science was anything more than an accumulation of observed facts. It should be borne in mind, however, that the Greek showed little sign of realizing the necessity of step-by-step progress. Throughout the whole course of Greek Natural Science we find a tendency to base a big generalization on a few observations; there is very little of that gradual building up of scientific theory that is characteristic of modern work, but there are constant attempts to pass from preliminary observation to final law of nature by great flights of the imagination. Science in its early days w~as afflicted by that common infirmity of youth, the desire to hurry on to the goal by short cuts ; it was not until age and experience added caution and hard work to the brilliance and enthusiasm of youth that any lasting progress was made. It seems probable, how- ever, that the main cause of the ultimate sterility of Greek Science was not so much rashness in the formulation of hypotheses as failure to check hypotheses by means of further observation or experiment. It is not strictly true to say that the test experiment was absolutely unknown, but it is certain that it was rare ; in its absence hypotheses were apt to become extravagant, and, moreover, their authors were forced back on to metaphysical arguments in defence of them. Thus metaphysics gradually expelled experience from scientific work till, in the Middle Ages, Science was practically smothered under a mass of meaning- less verbiage which had little relationship to reality. Yet while realizing this the modern scientist should be cautious in his criticism of his . predecessors in ancient or mediaeval times. It is easy nowadays to laugh at the old idea that the circle is the only perfect curve and therefore the paths of the planets must be described in PRESIDENTIAL ADDRESS PRIESTLEY. 5 terms of circular motion ; an idea from which even Kepler found it hard to break away and which hindered the development of Astronomy ; but it is not a great time since Biology and Geology were handicapped by a universal belief in the literal truth of the early chapters of Genesis, and at the present day Einstein is assailed with arguments based on the assumption of the objective reality of the iEther. The reaction against the barren metaphysical science of the Middle Ages set in towards the end of the sixteenth century and gradually gathered strength. The state of affairs in the early eighteenth century is well shewn up in the editor’s preface to the second edition of Newton’s Principia, published in 1713. In this preface Cotes, Plumian Professor of Astronomy and Experimental Phil- osophy in the University of Cambridge, summarizes the methods of the old and new schools of thought, and refers in addition to a third school occupying an intermediate position. That the new views had not found universal acceptance is evident from the fact that he considers it advisable to devote considerable space to a refutation of the Cartesian Yortex Hypothesis. In view of the light thrown on the scientific thought of Newton’s day by this preface, I ask your attention to a somewhat free translation of some extracts. " Students of Physics can be divided into three classes. For there are some who attribute specific and occult qualities to the several species of things from whence they derive, I know not by what process of reasoning, the behaviour of individual bodies. Herein is found the whole doctrine of the schoolmen, derived from Aristotle and the Peripatetics; they assert that each individual effect arises from the special natures of bodies f the origin of those natures they do not tell and hence they teach us nothing. Since all is in the names of things, not in the things them- selves, we can allow that they have found a certain philosophical language, but they have given us no Philosophy. "Others, again, have hoped to gain credit for better discretion by rejecting this useless accumulation of words. They maintain that there is a universal, homogeneous matter and that all variety of types which is observable in bodies arises from the simplest and easily understood relationships 6 PROCEEDINGS OF T«HE ROYAL SOCIETY OF QUEENSLAND. of their component particles. Indeed a certain advance is made here, passing from simpler to more complex phe- nomena, if to these particles they attribute no relationships other than those shewn in Nature. But when they take on themselves to postulate arbitrary unknown shapes and sizes and doubtful positions and motions, and further to invent some occult fluid, freely permeating bodies, endowed with perfect mobility and disturbed by occult motions; then are they given over to dreams, to the neglect of the real nature of things which, truly, is sought in vain by fallacious guesses seeing that it can with difficulty be studied by means of the most careful observations. Those who base their investigations on hypothesis, even if they then proceed most accurately according to the Laws of Mechanics, may be said to concoct fables, neat perhaps and beautiful but fables nevertheless. “ There remains the third school, which concerns itself with Experimental Philosophy. Its members attempt to derive the causes of all things from the simplest principles possible but they admit as a principle nothing that is not given by observed phenomena. They do not invent hypotheses and introduce them into Physics except as propositions into whose truth they must inquire. Their method falls into two divisions, one Analytic the other Synthetic. The forces of Nature and the simpler laws of forces are deduced by analysis of certain selected phe- nomena, and from them is found by synthesis ’ the nature of the remainder.7'' The method outlined in the last paragraph shews the main characteristics of modern scientific method ; the selection of the more outstanding results of observation for detached analysis, the formulation of a provisional hypothesis as a possible explanation the validity of which is to be tested further, the application of results obtained from the simpler problems in the building up of a general theory covering more complex eases, are all in keeping with our present-day views. It shews a marked advance on the earlier method of attempting to pass in one bound from observation to general theory. The need to proceed carefully to the hypothesis after preliminary analysis and selection of data had been shown by the failure of the old ways; the possibility of so doing PRESIDENTIAL ADDRESS PRIESTLEY. 7 had come with the improvement in means of observation and in the development of mathematical knowledge. An example of the importance of this progress is found in Kepler ’s work. Kepler’s laws of planetary motion arose from his study of the observations of Mars recorded by Tycho Brahe, in whose observatory he was an assistant. At the beginning of the work, under the sway of the old idea of the perfect circular motion, he attempted to describe the motion of the planet by means of epicyclics and very nearly succeeded. In his own account of his investigations we read: — “ Since the divine goodness lias given to us Tycho Brahe, a most careful observer, from whose observations the error of 8' is shown in this calculation, it is right that we should with gratitude recognize and make use of this gift of God. For if I could have treated 8' of longitude as negligible I should have already corrected sufficiently the hypothesis discovered in Chapter XVI ; but as they could not be neglected, these 8' alone have led the way towards the complete reformation of Astronomy.” Thus Kepler was wrarned off his epieyclic hypothesis by a discrepancy of eight minutes between calculated and observed results, an amount that could not have been detected by any Greek observer. Furthermore, his final explanation was in terms of the ellipse, a curve discovered by the Greek geometers, certainly, but unknown before metaphysics had obtained its hold on Greek Science. The fact that he arrived at a law that could not have been obtained by a Greek astronomer does not neces- sarily imply the superiority of his method, but may have arisen from the superiority of the means at his disposal. The detailed analysis of observations involved in the new scientific method required much description of phenomena and thereby caused an all-important change in the viewpoint of the scientific world. The question “Why” had dominated the older thought, but emphasis now was placed on the question “How.” It is unlikely that Newton and his contemporaries had adopted our modern view that the business of Science is to give an ordered description of phenomena and has no concern with first causes; but it is certain that they realized that 8 PROCEEDINGS OF TPIE ROYAL SOCIETY OF QUEENSLAND. adequate ordered description must provide the 'starting point of any further investigation. Some passages in Newton’s Principia have a distinctly modern ring; in particular the notes on the definitions show that many terms are intended to embody brief descriptions and carry no physical significance. For example, consider this passage : — “I use the terms ‘ attraction’ and ‘impulse’ in the sense of ‘source of acceleration’ and ‘source of motion.’ I use ‘attraction,’ ‘impulse,’ ‘tendency towards a centre,’ indifferently and as mutually interchangeable; these are to be understood not as physical but as mathematical con- cepts. Hence, let the reader beware of thinking that when I speak of ‘the attraction of a centre’ or ‘central forces’ I mean to imply any particular mode of action, or physical cause, or to attribute real physical powers to the centres (which are mathematical points).” This quotation provides the key to the whole scheme of the Principia. The first two books are concerned with a mathematical description of a certain type of motion, and the third book discusses various problems in connec- tion with the motion of the planets, moon and comets, tidal phenomena, and the precession of the equinoxes. These are all explained in terms of gravitational forces corresponding to the “attractions” of the first two books. Thus the whole work is devoted to the investigation of how our system moves. To extend the inquiry to the cause of this motion, it would first be necessary to pass from gravity as a convenient name to denote a source of acceleration towards a material particle, to gravity as an agency causing that acceleration by definitely explained means. Newton saw that this transition could not be made by means of further study of his phenomena, and, true to the newT ideals of science, he refused to make it in any other way. The final Scholium to the Principia contains this passage : — “I have not been able to deduce the reason of these properties of Gravity from the phenomena and I frame no Hypothesis. For whatever is not deducible from phenomena must be called hypothesis ; and hypotheses whether metaphysical, or physical, or of occult qualities, or mechanical, have no place in Experimental Philosophy. PRESIDENTIAL ADDRESS PRIESTLEY. 9 In this Philosophy propositions are to be deduced from phenomena and generalized by induction. It is sufficient for us that Gravity exists and, acting according to the laws we have put forward, is adequate to explain the motions of the heavenly bodies and the sea.” I have given these extracts from the Principia to illustrate the change of opinion as to the function of Science. The change of method is brought out more clearly by a consideration of the history of the origin of Newton’s great work. Tycho Brahe had accumulated by observation a mass of data on the positions of the planets ; Kepler, starting from the hypothesis that the path of Mars is an epicyclic, failed to account for observed facts and formulated the elliptic hypothesis. This covered the facts as far as Mars was concerned, so he extended it by a tentative assumption that all the planetary orbits are ellipses with the sun at one focus. This generalization was found to fit the facts and was enunciated in 1609 as Kepler’s First Law. The second law appeared in the same year and the third ten years later. In 1638 Galileo pub- lished his Dialogues on Two New Sciences, which contained the results of his work on falling bodies. The immediate result was the concentration of the scientific world on the problem of Gravitation. Halley, Wren, Huygens, and Plooke all attacked the subject. * Assuming tentatively that the planets were kept in their elliptic orbits by a force of the same nature as that causing bodies to fall to the earth, and, simplifying their problem by assuming circular in place of elliptic orbits, they deduced from Kepler’s Third Law that the attraction of the sun or earth on an external body must vary inversely as the square of the distance between the attracted and attracting masses. In passing we might notice here two characteristics of the new method; the attempt to find a single explanation of apparently different but possibly related phenomena ; and the simplification of the problem by ignoring temporarily certain of the data, in this par- ticular case the ellipticity of the orbit. The first is in accord- ance with the accepted “Rule, ” which was afterwards enunciated in the Principia, “No more natural causes are to be admitted than are sufficient to account for the phe- nomena”; the second follows the practice adopted by 10 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Kepler in concentrating on one planet. Mars, before attempting to describe the motion of all. The law of attraction derived from the simplified problem could be nothing more than a tentative suggestion. This was realized by its authors, who attempted to establish it more firmly by mathematical proof that under such a law and such a law only could Kepler’s Laws be true. In this attempt they evidently failed, for in August 1684 Halley visited Cambridge to consult Newton on the subject. The particular question to which he wanted an ansAver was, 4 4 What is the path of a planet under the inverse square law?” NeAvton had already attacked the problem and promised to send him a proof of the fact, discovered in 1679, that the path is an ellipse. On receiving this docu- ment in NoArember, Halley was so impressed Avith its importance that he made a special journey to Cambridge to persuade NeAvton to attack the Avhole problem of gravi- tation and to publish the results. The outcome was the publication of the Principia in 1687. I have given you this brief summary of the history of the Principia because it proAddes an excellent illustration of scientific method. We might notice again the main features : the careful accumulation of data by Tycho Brahe folloAvecl by the analysis and concise summary of the facts by Kepler ; the realization of the possible connection between the fall of bodies to the earth and the planetary motions; the framing of a tentative hypothesis suggested by a simplified problem, and finally a verification of the hypothesis by Avorking out its implications and comparing them Avith the results of obseiwation. Before leaving this (page of the history of Science, I would call your attention to the part played by Halley. Edmund Halley is generally known from his prediction of the return of the comet which bears his name ; his great claim to the gratitude of the scientific Avorld lies in his labours in connection with the Principia. Newton Avas notoriously reluctant to publish his work, and it is to Halley that the Principia oAves its existence, as is shown in Newton’s oavh preface — ‘ ‘ In producing this Edmund Halley, that man of great intellect and learning, laboured most earnestly. Not only did he correct the proofs and get the type engraved, PRESIDENTIAL ADDRESS PRIESTLEY. 11 but lie was the originator of the whole work. For when I had shown him that I had discovered the nature of the motion of the celestial bodies he never ceased to ask me to •communicate it to the Royal Society ; till at last he succeeded by his importunity and kindly encouragement in making me think of publication.” We know from other sources that Halley bore a large part of the cost of publication and dropped his own researches for a year or two in order to keep Newton up to the mark and push the great work through the press. The world owes much to the genius of Newton, but it owes no less to the perseverance, generosity, and self-sacrifice of Halley. Two facts stand out in the above outline of the development of scientific thought : the sterility of mediaeval science under a weight of metaphysical speculation, and the fertility restored by a return to experience which in less than a century produced the Principia. In view of these facts it was natural that subsequent developments should be characterized by a distrust of metaphysics. In the light of recent research, however, it appears that a wrong attitude to metaphysical questions may again lead to the stagnation of scientific work. We noticed, earlier in the evening, that one difference between ancient and modern science is the difference of aim; the old search after first causes has been replaced by attempts to formulate ordered descriptions. We no longer accept Newton’s rule and reason that “no more causes are to be admitted than are sufficient to explain the phenomena, because, as the philosophers say, Nature does nothing in vain and it is vain to use many means when few will suffice. Nature is simple and does not run riot with superfluous causes.” Rather do we accept the dictum of the French philosopher who said that Nature pays no heed to the difficulties of analysis, while at the same time we attempt to summarize our description of Nature in as few and as general state- ments as possible. Since our aim is description we are not concerned with the metaphysical problem of reality. The Newtonian scheme of Space and Time forms an adequate framework for the Newtonian Mechanics in terms of which has been built up a system of Natural Philosophy that has served us for two hundred years; the question of the 12 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. objective reality of Space and Time, as explained in the well-known and oft-quoted Scholium, does not affect the adequacy of this system as a means of description. Again, the iEther provides a convenient means of describing electromagnetic phenomena, but the description is equally satisfactory whether we regard the iEther as a mental concept or, alternatively, as an entity having real objective existence. It is certainly true that Natural Science as such has no concern with Metaphysics, but we must take care to give the correct meaning to the assertion. Metaphysical questions may be safely ignored as unsolved but irrelevant ; it is dangerous to avoid them by a tacit and unconscious assumption of a particular solution. A besetting sin of the scientific worker is a tendency to assume a realist solution of metaphysical problems; and the tendency is all the more dangerous in that it is unconscious. Consider, for example, the idea of space. By a process of abstraction from observations of material bodies the idea is developed and refined until we have the full concept of a three-dimensional continuum, subject to the laws of Euclidean Geometry, in which material bodies exist. By a metaphysical assumption this is endowed with objective reality and immediately certain .consequences follow; among others, displacements are combined by the parallelo- gram law and length becomes an intrinsic absolute property of material bodies. These are spoken of as facts of experi- ence; whereas they are actually the outcome of experience combined with certain metaphysical assumptions. Our direct experience indicates that length is a relation between object and observer which changes with change of relative position, but we introduce our assumptions about space and, by correcting for change of position, attribute absolute length to the object. Now nothing is gained by this tendency towards realism ; absolute length as a fact of experience is of no more value in scientific work than absolute length as a convenient interpretation of experience. On the other hand much flexibility is lost ; if we accept absolute length as fact we are limited in our interpretation of further experience, if we- look upon it as interpretation we can modify that interpretation as further experience demands. PRESIDENTIAL ADDRESS PRIESTLEY. 13 The tendency towards realism, which is all too common in the scientific world, arises mainly from the concentra- tion of individual workers on their own fields at the expense of due consideration of general scientific develop- ment. The scientist should take the advice so often tendered to politicians and social reformers — Bead history ! A thoughtful study of the history of the growth of science reveals the arbitrary nature of our current views, which have been adopted for convenience, not of necessity. Since all the sciences are influenced by Physics, it is sufficient to notice the way in which physical science- develops by mutual reaction between mathematical theory and physical observation. Observation suggests postulates for mathe- matical theory, and deductions from these postulates are tested by further observations. In order to carry out the tests it is necessary both to attach physical significance to the concepts of the mathematical theory, as Newton did when he identified a physical force of gravity with his mathematically defined central forces, and also to decide on canons of interpretation of the observations. The arbitrary element in the latter step is apt to be overlooked ; in fact, great man as Helmholtz was, he failed entirely to recognize it when he undertook his experiment to deter- mine whether space was subject to the laws of Euclidean Geometry. It was left to Poincare to point out that Helm- holtz ’s result was susceptible of two interpretations : space is Euclidean and a beam of light is straight, or space is non-Euclidean and a beam of light is curved. Further, it could not be upheld that the first interpretation must be taken on the grounds that Physics had already shown that light travelled in straight lines, for this fact was not a necessary conclusion from experience but merely a possible interpretation of experience. To return to the main question, when physical significance has been given to the mathematical theory, and when canons of interpre- tation of observations have been adopted, comparison can be made between theoretical and experimental results. Discrepancies between the two are reconciled by modifying the mathematical theory, the physical interpretation of the theory, or the interpretation of observations. By the constant interaction of theory and experiment a consistent 14- PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. scheme is built up, but at no stage is any step made of necessity; rather is it, in Poincare’s words, "the product of unconscious opportunism.” The final test of the adequacy of the scheme as a description of the physical universe is consistency. When we have shown this consistency we have a theory which is sufficient to account for the phenomena; to deduce the necessity of the theory we should have to prove the fact that no other consistent scheme could be evolved. Thus Science establishes the sufficiency but not the necessity of its description of Nature. The common tendency is, however, to treat the sufficient explanation as a necessary description and then to materialize the concepts embodied in it. We have seen that the revival of Science was accom- panied by a change of view as to its functions. The search for first causes gave place to an attempt to formulate ordered descriptions. There is need now to realize that these descriptions are not accurate w’ord pictures, but merely schematic representations, probably much simpli- fied, of underlying realities. As an example of the hindrance to the progress of Science which arises from the tendency to assume one particular solution of the meta- physical problem of reality, consider the history of the Theory of Relativity. In the last ten years Relativity has opened up a field of investigation which promises to be exceedingly fruitful. The experimental work from which the theory arises is over thirty years old, and the essential parts ot the mathe- matical theory by which it is "worked out have been in existence during the whole of that period. The delay of thirty years is due primarily to a failure to criticise the metaphysical assumptions underlying the Michelson- Morley experiment. The theory of that experiment rested on three assumptions — the reality of the iEther, and the absolute meaning of the terms "length” and "period of duration” as applied to objects and events; the last two being practically equivalent to the assumptions of the reality and mutual independence of space and time. When the experiment failed in its object, the determina- tion of the velocity of the earth through the iEther, the scientific world turned its attention to explaining away the failure by modifying the properties of matter and ignored PRESIDENTIAL ADDRESS PRIESTLEY. 15 the possibility of error in the metaphysical basis of the theory. It can be urged with some truth that chis attempt resulted in valuable scientific progress. My point is that with the solution of the metaphysical problem left open there were two possible lines of advance, that actually followed at the time and that adopted thirty years later. The assumption of a particular solution closed one path; it is true that the path has been reopened, but the reopen- ing was possible only after philosophical criticism had been introduced. The moral is obvious. The tacit assumption of the answers to underlying metaphysical questions, restricts the possible paths of scientific progress ; to reach its full development Science must invoke the aid of Philosophy. A certain type of scientist would counter this state- ment with the assertion that he makes no metaphysical assumptions ; he is concerned not with metaphysical reality but with physical reality of which the criterion is the possibility of measurement. This reply suggests Johnson’s famous refutation of Berkeley’s philosophy. The actual observations made in a scientific measurement are observa- tions of coincidences; the completion of the process of measurement consists in the interpretation of these observations, and the interpretation involves metaphysics. It would be foolish and unnecessary to demand that the scientist should solve the metaphysical problems, but he should at least recognize when he assumes solutions. This somewhat cursory survey of the history of scientific thought reveals two main xjeriods : the first characterized by crude metaphysical speculation and com- paratively barren; the second dominated by a return to experience, fruitful, but restricted in outlook by uncon- scious metaphysical assumptions. The pre -relativity work of Mach, Poincare, and others, and the general interest in the foundations of science that has accompanied the Theory of Relativity suggest that we are entering on a third period in which careful experiment will be combined with sound philosophical criticism. If this be so we can anticipate a period no less fruitful than its predecessor, and characterized by a breadth of view which in the past has too often been lacking. 16 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. On a Tertiary Fossil Insect Wing from Queensland ( Homoptera Fulgoroidea), with description of a New Genus and Species. By B. J. Tillyard, M.A., Sc.D. (Cantab.), D.Sc. (Sydney), C.M.Z.S., F.L.S., F.E.S., Entomologist and Chief of the Biological Department, Cawthron Institute, Nelson, N.Z. (Plate I and two Text- figures.) {Bead before . Royal Society of Queensland, 30th April, 1923) The beautiful fossil insect wing which forms the subject of this paper was discovered near Goodna, Q., by Mr. W. H. Bryan, M.Sc., Lecturer in Geology at the University of Queensland. Mr. Bryan, in sending me the fossil for descrip- tion, wrote as follows : — “ The specimen was collected by me from the Tertiary beds at Bedbank Plains, near Goodna, at the same spot and from the same horizon as your Euporismites halli, described on pp. 44-45, Queensland Geol. Survey Publ. No. 253, and figured on Plate 3. Associated with these wings are a fairly rich fish fauna and a number of well-pre- served dicotyledonous plants.” With regard to the age of the beds in which the fossil was found, there is some doubt, owing to lack of evidence of any Pleistocene glaciation and the absence of fossiliferous marine beds in the series ; but the presence immediately beneath these beds of vesicular trachyte, which Professor Bichards regards tentatively as belonging to his Middle Division of the Tertiary Volcanics of Queensland, suggests a Miocene age for the fossiliferous beds themselves. The fossil wing might at first sight be taken for one of the Psychopsid lacewings, owing to its great breadth, its general shape, and the density of its venation. But examination under a low power proves at once that it belongs to the family Bicaniidse of the Fulgoroid Homoptera, and is very closely allied to the recent Australian genus Scolypopa Stal, of which TERTIARY FOSSIL WING FROM QUEENSLAND — TILLYARD. 17 one species, S. australis (Walker), is the very common Passion- Vine Hopper of Eastern Australia, an insect very common throughout Queensland and the warmer parts of Hew South Wales. There is, indeed, no reason for not accepting the strong- probability that, in this new fossil find, we have actually a species which was the direct ancestor of our common Scolypopa. In order to facilitate comparison of the fossil and recent types, I have given in Text-fig. 1 a careful drawing of the actual fossil wing, which is practically complete except for the absence of the clavus or anal area, and in Text-fig. 2 a similar- drawing of the forewing of Scolypopa australis (Walker). The fossil requires a new genus for its reception, and I propose to name it Scolypopites bryani n.g. et sp., the generic name indicating its close affinity to Scolypopa , and the specific name being a dedication to its discoverer, who is to be heartily congratulated on his find. A comparison of the venational scheme of the two genera will be found attached to the generic definition. Order HEMXPTERA. Sub-order, HOMOPTERA. Superfamily EULGOROIDEA. Family RICANIIDiE. Genus SCOLYPOPITES n.g. Insects of the general build and facies of Scolypopa Stal, with very broad, closely- veined forewings. Venational scheme very similar to that of Scolypopa , but more primitive in the following characters (1) Sc. not so strong a vein as R, and only reaching to a little beyond the middle of the costal margin. (2) M and R not completely fused at bases. (3) Of the two gradate series of cross-veins found complete in Scolypopa , only the outer or marginal one is present in Scolypopites n.g. As in the Psychopsid and Osmylid Lacewings, this gradate series divides the wing into a central “ disc ” and an outer marginal area. In Scoly- popites n.g. there are numerous weakly formed and irregularly placed cross- veins within the disc. R.S. — C. 18 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Text-figure 1 : — -Scolypopites bryani n.g. et sp. Forewing (x 6). The missing clavus is restored by dotted lines. 1A, 2 A, the two anal veins, forming the claval Y-vein ; Cu, cubitus ; Cux, its upper branch ; Cu2, its lower branch, the vena dividens ; M, media, with its four main branches, Mx to M4 ; Rx, main stem of radius ; Rs, radial sector, branching into R2 + 3 and R4 + 5 ; Sc, subcosta. Tertiary (? Upper Miocene) of Goodna, Q. Text-figure 2 :—Scolypopa australis (Walker). Forewing (x 11, For comparison with Text-figure 1, which see for venational notation. (Actual size. 8 mm. long by 5 mm. wide). Recent, Eastern Australia. Proc. Roy, Soc. Q’land, Yol. XXXV, Plate I Face page 19. Scolypopites bryani, Tillyard, n.g, et sp. TERTIARY FOSSIL WING FROM QUEENSLAND TILL YARD. 19> In Scolypopa Stil, all these have been eliminated with the exception of a complete set which forms a second or discal gradate series, and one or two more basally situated between M and Cu. (4) In Scolypopites n.g., Its is strongly branched not far from its origin. In Scolypopa Stal, Rs. is either reduced to a simple vein, as in Text-fig. 2, or it does not branch until about midway along the wing. (5) The manner of branching of M and Cuj in Scolypopa is very variable. Text-fig. 2 is taken from a speci- men in which the branching of M. is trifurcate, as in the fossil. Many specimens, however, show the two veins M labelled M3 and M4 arising from a single stem which is itself a dichotomy with the vein marked M4+2 ; it is for this reason that these veins are so named. I think, therefore, that the trifurcate forking of M in Scolypopites n.g. may be only an individual character of this particular wing, and so I do not propose to include it in the generic definition. For the same reason I omit mention of the particular form of branching of Cu4 and its manner of connection with M4, since these also are highly variable in Scolypopa. Genotype : — Scolypopites bryani n. sp. Horizon : — Tertiary Beds (probably Upper Miocene) of Goodna, Queensland. Scolypopites bryani n. sp. (Plate I and Text-fig. 1.) The fossil consists of the tegmen or forewing only, well preserved, but with the clavus or anal area entirely absent, as is very usually the case with fossil Homoptera, owing to the deepness of the impression of the vena dividens (Cu2), which causes the wing membrane to split, so that the clavus usually drifts away from the rest of the wing and becomes fossilised by itself. Total length 15 mm. ; greatest breadth (measured from tornus to costa at right angles to the latter), 9;5 mm. The impression is on a dark ochreous-brown sandstone rock, rather 20 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. soft ; the outlines of most of the veins are distinct, but the impression is slightly blurred here and there. Dotted portions of Rj and Cu in Text-fig. 1 indicate restorations where the impression is not clearly visible. The principal specific characters are as follows : — The large size, the length and breadth being almost twice that of Scoly- popa australis , and very much larger than usual for the family. In the venation, the very strong basal arching of the costa, and its slight waviness beyond the end of Sc ; the less rounded apex compared with that of Scolypopa australis ; the greater number and closeness of the costal and marginal veinlets, and the larger number of forkings of most of the principal branches of the main veins ; also the weakness of the closure of the disc from the end of Sc to the top of the marginal gradate series near the apex. The clavus has been restored, in Text-fig. 1, much on the lines of that of Scolypopa australis Stal. Weak irregular cross- veins, little more than mere impressions on the membrane, are generally present in the clavus of this species, and may well have been present also in the clavus of the fossil, but it has not been thought wTorth while to indicate such in Text-fig. 1. For a full comparison of the two types of venation, see Text-figs. 1 and 2. Type — Unique holotype forewing, in Coll. Geol. Dept., University of Queensland. In conclusion, I wish to thank Mr. W. C. Davies, Curator of the Cawthron Institute, for the excellent photographic enlargement of the fossil wing from which the Plate has been prepared ; owing to the dark- brownish colour of the rock it proved by no means an easy task to photograph this specimen. Cawthron Institute, Nelson, N.Z., Dec. 11th, 1922. LIFE-CYCLE OF CERTAIN AUSTRALIAN BLOWFLIES. 21 Observations regarding the Life-Cycle of Certain Australian Blowflies. By Professor T. Harvey Johnston, M.A., D.Sc., University, Adelaide ; and G. H. Hardy, Walter and Eliza Hall Fellow in Economic Biology, University, Brisbane. (Read before, Royal Society of Queensland, 30th April, 19.23) Experiments in breeding blowflies previously carried out in the biological laboratory of the University, Brisbane, by Johnston and Tiegs, showed variations in regard to the duration of different stages in their life- cycle, and in order to collect additional data upon this subject similar flies have now been bred in larger numbers than previously. Decomposing meat was exposed for a period of from six to eight hours on certain selected days during the present summer (1922-2-3), so that flies in the neighbourhood might be attracted to the carrion and perhaps oviposit thereon. The material was then transferred to an insectarium and completely isolated from further infestation. Each clay such larvae as had left the meat to pupate and had buried themselves in the sand prepared for them were sifted out with a sieve made of mosquito -netting and then transferred to one or more vessels containing sand, wherein, they were allowed to pass their prepupal period. Each day these prepupae were again examined, and such as had become pupae were sifted out and transferred to the laboratory to await their emergence. A system of consecutively numbering each batch to larvae and pupae as they passed through their stages enabled us to keep individual records for every specimen handled. From such records the tables published in this paper have been compiled. Before starting these experiments various observations were made to ascertain whether the handling of prepupae or pupae in the above manner interfered with the various periods to any appreciable extent, but we could find no evidence for considering that it did. The conditions producing alterations in such periods as the season advanced were apparently climatic changes. 22 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. In this paper the term “ prepupa ” is applied to cover the maggot from the time of wandering away from the meat till it has begun to form a puparium, when further movement becomes impossible. There is reason to suppose, however, that some larvae remain in the meat for one, two, or even more days after having become fully fed, thus extending their calculated feeding period and diminishing the recorded prepupal period. A case in point may be mentioned relating to a batch of Sarcophaga beta J. & T., that was bred by us and given special attention. Four days after deposition all the maggots were fully grown ; nothing was left of their short food supply but slime, and five of the fifteen specimens left it to pupate, the others remaining in the slime. The next day three further specimens migrated, while the balance remained for eight more days before leaving and during the whole of this period scarcely moved their position and did not increase in size. If in the table given later in this paper, four days be allowed as the feeding period and the balance of the figures above this number be added To the prepupal period, then the latter would vary from 4 to 12 days, whilst the pupal period would vary not more than one day either side of the 14 days, this being a comparatively constant period, so that given the date of deposition and the date of emergence, the dates of other periods should be readily estimated with a reasonable degree of accuracy. Although this appears to apply to the spring and summer time, it does not hold good for the winter, when Sarcophaga pupae vary widely in their pupal period. We wish to emphasise the wide variation in the length of individual life- cycles, such being probably due mainly, if not entirely, to the varying time the insect takes to pass through the prepupal condition. The climbing abilities of blowfly maggots are great, and many prepupae will escape if special precautions be not taken. In three of the larger series we used a metal receptacle with an inwardly curved lip. and without cracks or crevices. When sifted out from this the larvae were then placed in a glass jar with a wide open mouth and with a shoulder sufficiently wide to again form an inwardly curved obstacle against the maggots’ climbing powers. Once the insect has become sufficiently far advanced in its prepupal condition (i.e. after not less than three days) further precautions against escape are unnecessary, provided the maggots are not subjected to damp, which may again induce them to wander, if pupation has not actually LIFE-CYCLE OF CERTAIN AUSTRALIAN BLOWFLIES. 23 commenced. The meat was placed in a bare shallow dish so that the larvie had to make a special effort to leave it, thus ensuring that none did so before they were fully fed. If the meat were placed on sand some maggots would attempt to pupate in the dish, and this would necessitate disturbing the feeding larvse in order to procure such pupse. Very rarely did maggots pupate in the meat, and then only if it were allowed to dry out. Table Indicating Duration of Various Stages. The tables below indicate a complete tabulation of individual specimens bred through from their deposition to emergence. All intermediate stages which failed to complete their development have been excluded ; in other words all larvae, prepupae, or pupae that died before emergence occurred, or were destroyed by parasitic agencies, have been omitted from this particular record. The first entry in the table is intended to record that one specimen, deposited by its parent on the 12th October, finished its feeding and left the meat in which it wTas being bred, on the 16th, so that its larval period was four days. This specimen pupated on the 18th, thus requiring a further two days to complete its pupation, i.e. the prepupal stage was two days and the period from deposition to pupation (total larval period) was six. By the 26th, the imago had emerged, the pupal period being eight days and the total time from deposition to emergence fourteen. The species of Lucilia used was the common blue or green bottle of Brisbane, and generally regarded as L. sericata . Tables 1 to 4 refer to it. No. of Specimens Bred. 24 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Date of- No. of Days From — o ■£ > a di-3 ^■3 PI Eh a a> c . £ O eS 'S & S SPH Ph_ Is Table 1. Lucilia sp. Oct. Oct. Oct. Oct. 1 12 16 18 26 4 1 6 14 2 8 6 12 16 19 26 4 7 14 3 7 109 12 16 19 27 4 7 15 3 8 6 12 17 19 27 5 7 15 , 2 8 1 12 18 19 27 6 | 7 15 ■ ! 1 8 2 12 16 20 27 4 i 8 ! 15 4 7 2 12 17 20 27 5 8 1 15 | 3 7 8 12 16 19 i 28 4 7 16 j 3 9 3 I 12 17 19 28 ! 5 7 16 2 9 39 12 16 20 28 4 ! 8 16 4 8 40 12 17 20 28 5 8 16 j 3 8 11 12 18 20 28 6 8 1 16 2 8 3 12 18 1 21 28 6 9 16 3 7 4 12 j 16 - 20 29 I 4 8 17 1 4 9 6 12 17 20 29 5 8 I 17 3 9 2 12 18 20 29 6 8 17 2 9 7 12 16 21 29 4 9 17 ! 5 8 29 12 i 17 21 29 5 9 17 4 8 79 12 ! 18 21 29 6 9 17 3 8 12 12 19 21 29 7 9 17 2 8 3 12 17 22 29 5 i 10 17 1 5 7 3 12 16 21 30 4 9 18 5 9 5 12 17 21 30 5 ! 9 18 4 9 10 12 18 21 30 6 9 18 l 3 9 3 12 19 21 30 7 1 9 18 2 9 1 12 ! 20 21 30 8 9 18 1 9 2 12 16 22 30 4 ! io 18 ; 6 8 3 12 17 22 30 5 10 18 1 5 8 60 12 18 22 30 6 i 10 18 4 8 41 12 19 22 30 ! 7 10 18 3 8 1 12 20 22 30 8 i 10 18 2 ! 8 1 12 17 22 31 5 10 19 5 9 17 12 18 22 31 6 10 19 4 9 8 12 19 22 31 7 10 i 19 3 9 1 12 20 22 31 8 i 10 19 1 2 9 5 12 17 23 31 5 11 ! 19 6 8 26 1 12 j 18 23 31 6 11 19 5 8 19 12 | 19 23 31 7 11 19 4 1 8 8 12 1 20 23 31 8 11 19 3 | 8 No. oi Specimens Bred AFE-CYCLE OF CERTAIN AUSTRALIAN BLOWFLIES, 25 1 Oct. 12 Oct. 21 Oct. 23 Oct. 31 9 11 i 19 2 8 1 12 18 24 31 6 12 19 6 7 1 12 20 24 31 8 12 19 4 7 1 12 21 24 31 9 12 1 19 3 7 1 12 17 23 Nov. 1 5 11 ■ 20 6 9 1 12 18 23 1 6 11 20 5 9 6 12 19 23 1 7 11 20 4 9 1 12 17 24 1 5 12 20 7 8 27 12 18 24 1 6 12 20 6 8 7 12 1 19 24 1 7 12 20 5 8 6 12 20 24 1 8 12 20 4 8 '2 12 21 24 1 9 12 20 3 8 1 12 18 23 2 6 11 21 5 10 1 12 18 24 2 6 12 21 6 9, 2 12 19 24 2 7 12 21 5 9 18 12 18 25 2 6 13 21 7 8 14 12 19 25 2 7 13 21 6 8 2 12 20 25 2 8 ! ! 13 21 5 8 1 12 19 24 3 7 1 12 | 22 5 10 7 12 18 25 3 6 13 22 7 9 3 12 19 25 3 7 13 ! 22 6 9 1 12 16 26 3 4 14 22 10 8 5 12 18 .26 3 6 14 22 8 8 8 12 19 26 3 7 14 1 22 7 8 3 12 17 26 4 5 14 23 9 9 5 12 18 26 4 6 14 23 8 9 6 12 18 27 4 6 15 1 23 9 8 4 12 19 27 4 7 15 | 23 8 8 1 12 20 27 4 8 15 23 7 8 1 12 22 27 4 10 15 23 5 8 3 12 19 26 5 7 14 24 7 10 1 12 22 26 5 10 14 24 4 10 2 12 19 27 5 7 15 i ! 24 8 9 4 12 19 28 5 7 16 ! ! 24 9 8 10 12 18 28 5 6 16 24 10 8 2 12 16 28 5 4 16 24 12 8 1 12 20 28 6 8 16 25 8 9 2 12 19 28 6 7 16 25 9 9 3 12 18 28 6 6 16 25 10 9 1 12 17 29 6 5 17 25 12 8 9 1 12 18 29 6 6 17 25 11 8 No. of Specimens Bred, 26 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND, 5 1 2 1 13 3 3 2 6 3 1 2 l 1 1 1 1 1 Date OF — No. of Days From — Deposition. ! Prepupation. Pi o trt ft PM Emergence. [ Deposition to Prepupation = Larva Feeding. Deposition to Pupation = Total Larval. Deposition to Emergence = Total Period. i Prepupation to Pupation = Prepupal Period. 1 Pupation to Emergence = Pupal Period. T A BLE 1 — -contim ted. o O Oct. 1 Oct. | Nov. i 12 19 29 | 6 7 17 25 10 8 12 ' 21 20 6 1 9 17 25 8 8 12 IS 29 7 6 ; 17 26 11 9 12 20 29 7 8 17 26 9 9 12 IS 30 7 6 18 26 12 8 12 19 30 ' 7 ! 7 18 26 11 8 12 18 30 8 6 18 j 27 12 9 12 19 30 8 7 18 27 11 9 12 18 31 8 6 19 27 13 8 12 19 31 8 7 19 27 12 8 Nov. 12 18 1 9 6 20 28 14 8 12 19 1 9 7 20 28 13 8 12 19 2 10 7 21 29 14 8 12 20 2 10 8 21 29 13 8 12 19 3 11 7 22 30 15 8 12 19 4 12 n i 23 ! 31 | 16 8 12 19 5 12 7 24 31 17 7 1 12 I 19 1 5 13 7 ! 24 l 32 17 8 TABLE 2. Lucilia sp. 42 Nov. 2 1 Nov. ■ 7 Nov. 10 Nov. 17 5 8 15 3 7 1 2 8 10 17 6 8 15 2 7 2 2 9 10 17 7 8 15 1 7 45 2 7 10 18 5 8 • 16 3 8 2 2 8 10 18 6 8 16 2 8 3 2 9 10 18 7 8 16 1 8 147 2 7 11 18 5 9 16 4 7 7 2 8 11 18 6 9 16 I 3 7 5 2 9 11 1.8 7 9 16 2 7 2 1 2 7 12 18 5 10 16 5 6 4 I 2 ' 7 10 19 5 8 17 3 9 69 j 2 ; : 7 11 19 5 9 17 4 8 27 2 8 11 19 6 9 17 3 8 2 i 2 : 9 11 19 7 9 17 2 8 110 : 2 i 7 12 1 19 5 10 , 17 ! 5 7 39 1 2 1 8 12 19 6 10 17 4 7 24 : 2 9 12 19 7 10 17 i 3 7 No. of Specimens Bred. LIFE-CYCLE OF CERTAIN AUSTRALIAN BLOWFLIES. 27 Date of — No. of Days From — Table 2 — continued. Nov. Nov. Nov. Nov. 61 2 7 12 20 5 10 j 18 5 48 2 8 12 20 6 10 18 4 29 2 9 12 20 7 10 18 3 4 2 10 12 20 8 10 18 2 2 2 11 12 ' 20 9 10 18 1 51 2 7 13 20 5 11 18 6 56 2 8 13 20 6 | 11 18 5 12 2 9 13 20 7 11 1 18 4 2 2 10 13 20 8 1.1 18 O O 1 2 11 13 20 9 11 18 2 9 2 7 12 21 5 10 j 19 5 30 2 7 13 21 5 11 19 6 59 2 8 13 21 6 11 19 '5 18 2 9 13 21 7 11 19 4 2 2 11 13 21 9 11 19 2 22 ■2 7 14 22 5 12 19 7 88 | 2 8 14 22 6 I 12 19 6 12 j 2 9 14 22 7 12 19 5 1 2 10 14 22 8 ■ 12 19 4 3 2 11 14 22 9 12 19 3 4 2 ’ 7 14 22 5 12 20 7 12 2 8 14 22 6 12 20 6 2 2 9 14 22 7 12 20 5 1 2 11 14 22 9 12 20 3 4 2 7 15 22 5 IS 20 ■ 8 105 2 8 15 22 | 6 13 20 7 5 2 9 15 22 1 7 13 20 6 4 1 2 11 15 22 9 13 20 4 3 ! 2' 8 16 22 6 14 20 8 1 2 12 15 23 10 13 21 3 63 2 8 16' 23 6 14 21 8 3 2 9 16 23 7 14 21 7 3 2 , 11 16 23 9 14 1 21 5 1 2 13 16 23 11 14 21 3 2 ■ 2 8 16 24 6 14 22 8 2 2 13 16 24 11 14 22 3 1 ! 2 14 16 24 12 14 22 ' 2 17 t 2 8 17 24 6 15 22 9 5 1 2 8 17 25 6 15 23 9 1 1 2 i 8 18 25 6 16 23 10 1 2 8 18 27 6 16 25 10 1 2 ! 8 19 27 6 17 25 11 4CcooGc > G .2 ~G ’ G g ' 0) o d ' W>_T f-i T3 a> o G . o |-£j P O G p ft 03 G p O ft G ft o Stj bDO 2 03 CP P_ _ G 02 ft m fl o +3 O '■£ a o . May Mav May 1 10 15 28 5 13 1 10 15 Aug. 5 105 Jun . Jun. Jun. 26 1 2 20 23 27 18 21 86 3 65 Sep. 1 2 20 23 13 18 21 103 3 82 1 2 20 23 14 18 21 104 3 83 1 2 20 23 26 18 21 116 3 95 1 2 20 23 27 18 21 117 3 96 1 | 2 20 29 28 18 27 118 9 91 Oct. 1 2 23 1 . 4 21 124 106 froggalti Tayl. Jan. Jan. Jan. Feb. 3 18 26 30 9 8 12 22 4 10 4 18 ; 26 31 10 8 13 23 5 10 2 18 26 31 11 8 13 24 5 11 The information given above constitutes the first published data regarding the life-cycle of S. misera, S. eta, and S. froqqatti (Syn. SI theta J. & T.) The observations regarding S. peregrina assist in filling certain gaps in the information contained in Johnston and Tiegs’ paper. LIFE-CYCLE OF CERTAIN AUSTRALIAN BLOWFLIES, 41 TABLE 10. 1 Oct. 12 Oct. 22 Oct. 28 Nov. 7 10 16 26 6 10 12 12 29 7 17 26 9 6 12 29 8 17 27 10 5 12 30 8 18 27 9 12 12 30 9 18 28 10 1 12 31 9 19 28 9 4 12 31 10 19 29 10 1 12 22 31 10 10 19 29 9 10 5 12 Nov. 1 10 20 29 9 1 12 22 2 10 10 21 29 11 8 1 12 22 1 11 10 20 30 10 10 2 12 1 11 20 30 10 1 12 22 2 11 10 21 30 11 9 1 12 2 11 21 30 9 4 12 22 2 12 10 21 31 11 10 2 12 2 12 21 31 10 5 12 22 3 12 10 22 31 12 9 2 12 3 12 22 31 9 1 12 22 3 13 10 22 32 12 10 4 12 3 13 22 32 10 1 12 20 4 13 8 23 32 15 9 2 12 4 13 23 32 9 6 12 20 4 14 8 23 33 15 10 2 12 4 14 23 33 10 1 12 4 15 23 34 11 1 12 22 6 15 10 25 34 15 9 2 12 6 15 25 34 9 4 12 22 6 16 10 25 35 15 10 1 12 6 16 25 35 10 1 12 22 7 16 10 26 35 16 9 4 12 22 7 17 10 26 36 16 10 1 12 22 8 18 10 27 37 17 10 1 12 13 24 32 43 11 42 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. This set of observations relates to that period regarding which Johnston and Tiegs’ paper shows certain gaps. The larval feeding stage during October was 8 to 10 days, the prepupal 6 to 17, and the pupal (October-November) 8 to 10, while the total period elapsing between deposition and emergence was found to range from 26 to 43 days. LITERATURE REFERRED TO. 1922. Johnston and Tiegs. — Notes on the biology of some of the more common Queensland Muscoid Flies. Proc. Roy. Soc. Queensland, xxxiv.. pp. 77-104. A CORRECTION. In our previous paper, these Proceedings, vol. xxxiv, 1922, the following lines should be erased: — - Page 192, last line Page 193, first line. CACTUS BUGS OF GENUS ■ CHEL1NIDEA (HEMIPTERA). 43 New Cactus Bugs of the Genus Chelinidea (Hemiptera) By John C. Hamlin, M.Sc., Officer i* Charge, Prickly-pear Investigations, Commonwealth Prickly-pear Board. {Read before the Royal Society of Queensland , 28th May , 1923.) The cactus bugs of the genus Chelinidea form one of the important groups of insects which are being introduced in the effort to control prickly-pear in Australia by biological agencies. During the past two years the writer has collected these insects extensively in North America with the result that two un- described species and a new variety have been discovered. The new forms have been referred to in our work under the writer’s manuscript names, but, in view of their economic status, it is highly desirable that their descriptions should be published. Chelinidea hunteri n. sp. Head subequal in length to pronotum, ocelli nearer to eyes than to pronotal collar, juga rather abruptly pointed and a very little exceeded by the tylus. Pronotum more con vexed than in C. vittigera, lateral margins nearly straight from humeral angles to collar, anterior margin without teeth. Connexivum little dilated, extending only as a narrow edge beyond the hemelytra ; hemelytra and connexivum forming straight parallel lines. Under surface of fore femora with from three to six teeth distally, arranged in two rowr Tibiae merely carinate. Ground colour faded yellow ; head dark brownish or fuliginous with a paler indistinct vitta extending from over the tylus to base of the head, antennae fuliginous except distal joint; pronotum with fuliginous transverse vittae near front and rear margins ; coriaceous portion of hemelytra mottled with rusty brown and smoky areas, the veins pale and bordered by thin lines of black ; membrane blackish bronze ; beneath, uniform greenish yellow, with legs a shade darker, and tip of rostrum piceous 44 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Length 9-5 to 10*5 mm ; humeral breadth 3-5 to 4 mm. Described from four male specimens collected by the writer on August 7 th, 1922, at Ranchi to, near Hermosillo, Sonora, Mexico, feeding on a cylindropuntia of the Imbricatce series. Hunter et al. refers (Bur. Ent. Bull. 113, p. 20) to a small form found upon 0. arbuscula, 0. versicolor, and 0 . fulgida at Tuscon, Arizona (U.S.A.). There is little doubt that the above-described species is the same referred to by Hunter. The type material is in the writer’s collection. Chelinidea canyona n. sp. Head and pronotum subequal in length ; ocelli nearer to extremities of pronotal teeth than to eyes, juga rather abruptly pointed and just attaining the tip of the tylus or very slightly exceeded by it. Pronotum appearing slightly concave ; the lateral margins elevated, forming distinct, laterally com- pressed crests which are curved upward ; anterior margin with a strong acutely pointed tooth on either side. Connexivum greatly dilated and inclined upward. Under surface of fore femora bearing distally two or three small teeth. Tibiae triangular with all edges prominently elevated ; hind tibiae with the two outer edges almost foliaceous. Penultimate ventral segment of female medianly cleft to two-thirds of its width, the edges of the incision slightly rounded and barely overlapping ; margins either side of notch very slightly concave. Colour above rusty brown with darker markings. Head brown with a paler vitta on tylus extending to base of the head and bordered by shining black ; pronotum with a wide border of black just before the posterior margin, and very narrow ones along the lateral margins ; hemelytra with three dark bluish vittse each, one along the outer margin and two oblique ones converging with the first proximally ; membrane black ; connexivum dark with segments indicated by pale lines ; beneath yellowish except pectus, bases of legs, and tip of rostrum, shining black ; legs brownish yellow except coxse and bases of femora. Length 11 to 14 mm. ; humeral breadth 4 to 4-5 mm. Holotype, allotype, and paratypes are in the collection of the writer. The writer first collected this species in the Rio Frio canyon, near Rio Frio, Texas (U.S.A.), in June 1921. Subse- CACTUS BUGS OF GENUS CHELINIDEA (hEMIPTERA). 45 quently, it was taken generally over the canyon country north of Uvalde, Texas, but its distribution is apparently limited to the canyons. Its food is the prickly-pears of such regions. Since the above date I have constantly referred to it in my reports and correspondence under the manuscript name of Chelinidea canyona. Chelinidea vittigera Uhler, var. texana, n. var. This variety differs from the typical C. vittigera Uhler mainly in that the representatives are slightly less robust and that they lack the colour markings of the described form. The colour is uniformly testaceous, with membranous portion of hemelytra and tip of beak smoky black. Length 10 to 13 mm. ; humeral breadth 3 to 4-5 mm. It is this form of Chelinidea which is the common cactus bug of Texas. In that State I have taken it at Kingsville, Brownsville, Laredo, San Antonio, Uvalde, La Pryor, Con Can, and Eagle Pass. In Mexico I have taken it in the country just south of Piedras Negras (Coahuila) and at Monterey (Nuevo Leon). Besides the various Opuntias of the regions mentioned, I have occasionally found it feeding upon Echinocereus sp. (“ pitallo”) and Opuntia leptocaulis (“ tasajillo”). 46 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Some Further Observations on the Dawson River Barramundi : Sclero- pages leichhardtii. By Thos. L. Bancroft, M.B. (Read before the Royal Society of Queensland, 28th May, 1923.) Professor T. Harvey Johnston, on my behalf, communi- cated to the Society a few curtailed notes on the Barramundi on 29th May, 1916, 1 to which this paper is supplementary. The embryology of an ancient type of fish, such as Scleropages, is most important ; so that any observations as to methods of obtaining material in the proper state to submit to embryologists are worthy of record. Perusal of these methods would obviate much waste of time and be of the greatest value to future scientific explorers. It used to be said that Barramundi would not mesh, but I found that was because nets of large mesh were not used, for with a six- inch mesh it is quite easy to take large fish, and even with a four- inch mesh fish up to four pounds in weight are readily meshed. In my former paper I predicted that Barramundi might carry its ova in the mouth, and this has since been found to be so. Not being able myself to go again to the Dawson I induced some young fellows who have a selection at Mostowie, the Squire brothers, to try to secure Barramundi with ova in the mouth. I supplied them with nets of various-sized mesh, one with a very small mesh suitable for dragging. Unfortunately the Dawson River, and likewise a large billabong on their selection in which Barramundi abound, are so full of fallen trees and snags that dragging a net is impossible. With set nets they found it quite easy to mesh the fish, but invariably, owing to the struggles of the fish captured in this way, any ova in their mouths are ejected. The Squire brothers, however. 1 Proc. Roy. Soc. Q’land, vol. xxviii, p. 93. DAWSON RIVER BARRAMUNDI : SCLEROPAGES LEICHHARDTII. 47 on 21st October, 1922, shot with a rifle several Barramundi, the mouth of one of which was full of ova in various stages of development. The best method to follow in securing ova and young forms would be to find a suitable spot on the Dawson to drag a net ; a strong hemp net of three-inch mesh about fifty yards in length, with wings ten feet deep and a bunt eighteen feet, would, in my opinion, be the most suitable. The facts so far obtained might be shortly expressed thus : — 1. Barramundi carries the spawn in its mouth. 2. October is the spawning season. 3. Fish when meshed eject the spawn. 4. Sexually mature fish are not meshed in a net of three-inch mesh. 5. Fish secured in the bunt of a drag-net would retain the ova and young fish in their mouths. On 12th November, 1922, whilst dragging a net in the' Burnett River, a Salmon Catfish, Hexanematichthys australis , was taken with ova in its mouth. This fish is uncommon in the Burnett but extremely plentiful in the Dawson River ; the latter river would be the place to go if ova of this fish were required. 48 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. An Unusual Tourmaline-Albite Rock from Enoggera, Queensland. By W. H. Bryan, M.Sc., Lecturer in Geology and Mineralogy, The University of Queensland. (Plate II.) ( Read before the Royal Society of Queensland, 28th May, 1923.) (I) Introduction. (II) Field Occurrence. (III) Descriptive. (IV) Chemical. (V) Discussion — (а) Mineralogical. (б) Structural. (VI) Conclusion. Plate. (I) INTRODUCTION. The remarkable rock which forms the subject of this paper was discovered in the year 1914 by the author on the Cedar Creek road, between portions 313 and 314, parish of Kedron, about 7 miles north-west of the City of Brisbane, in the State of Queensland. The only previous record of this rock is a short reference in the author’s “ Geology and Petrology of the Enoggera Granite and Allied Intrusives, Part I ”.1 (II) FIELD OCCURRENCE. The rock to be described forms a band about two feet in thickness some 25 yards outside the contact of the “ Enog- gera Granite ” (of Permian age) with the Brisbane Schist (? Ordovician). This band can be traced for fifty yards in a 1 Proc. Roy. Soc. Qld., vol. xxvi, 1914, p. 153. UNUSUAL TO URM ALINE- ALBITE ROCK FROM ENOGGERA. 49 N.N.W. direction, which is the normal strike of the Brisbane Schists, and is, further, the most important trend line in the Brisbane area. This agreement in direction may, however, be merely coincidental, for the schists twenty yards further from the contact strike E.N.E. (i.e., roughly parallel with the edge of the granite laccolite at this point), and dip to the N.N.W. , while between the tourmaline- albite rock and the contact the strike is very erratic. The appearance of the rock in the field, s very beautiful and striking, being black in colour with numerous irregular sinuous and contorted veins of a light- coloured material standing out in marked relief. A few feet away on either side an identical structure can be seen where numerous tortuous veins of quartz are threaded through the normal mica schists.2 The field evidence leaves no doubt that, how- ever the mineralogical nature of the tourmaline- albite rock be explained, the schistose structure is that of the Brisbane Schists themselves. Nearer the granite and almost on the contact is a large body of massive schorl rock composed almost entirely of tourmaline identical with that in the tourmaline- albite rock but in which only traces of the schistosity remain. These traces sometimes take the shape of numerous irregular roughly parallel but empty veins. This rock contains, in addition, cavities which are partly or almost completely filled with well-shaped quartz crystals and an occasional crystal of pyrites. Small patches of tourmaline are also found in the schist over fifty yards from the contact. These tourmaline rocks cannot be considered as in any way representative of the contact effects of the Enoggera Granite on the schists.3 The Brisbane Schists outside the con- tact zone are in this locality a series of schistose and phyllitic rocks, consisting for the most part of mica, through which run numerous irregular veins of quartz, which follow the plica- tions and thus emphasize the schistose character of the rock. The normal effect of the intrusion of the granite at this point is to produce in these schists a recrystallization of both mica and quartz, but no new minerals seem to have been added. 2 See Richards, “ A Study of the Brisbane Schists,” Inst, of En gineers, Brisbane Divn., June 1922, p. 5. 3 See Bryan, 1914, op. cit., p. 152. R.S. — E 50 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. (Ill) DESCRIPTION OF ROCK. Megascopic. — In the hand specimen the rock appears to be made up entirely of black tourmaline and creamy pink felspar. The tourmaline, which consists of a matted aggregate of minute prismatic crystals, is the more abundant mineral, and forms a black background against which the light-coloured felspar stands out in high relief as a series of crumpled, con- torted bands of varying thickness. The individual bands are sometimes persistent, but are often pinched out and discon- tinuous. They vary in width from 2 mm. to barely discernible threads. Microscopic. — Microscopic inspection confirms the general impressions gained in an examination of the hand specimens. The rock is seen to consist almost entirely of tourmaline and felspar. The tourmaline forms a matted aggregate of prismatic crystals averaging about 0T5 mm. in length and 0 03 mm. in breadth, and of fairly uniform size. In transmitted light the colour is bluish green. Absorp- tion is strong and characteristic (0>E) and the dichroism is marked, O being dark bluish green, and E light brown with a greenish tint. All of the tourmaline crystals appear to be of the one type and no colour zoning is apparent in individuals. The felspar is albite, with typical double refraction and low refractive index. Polysynthetic twinning is commonly present both on the Albite and Pericline law, both types being sometimes displayed on the one crystal. The twinning is somewhat irregularly developed and sometimes gives rise to a structure which may be compared with the “ Chequer ” structure of albite described by Flett,4 Hughes,5 and Tilley.6 The felspar crystals contain numerous minute inclusions. The junction between the tourmaline and felspar is well defined. A few crystals project into the felspar, and occasional isolated crystals are seen enclosed in the felspar itself, but the general nature of the contact militates against the view that the tourmaline has been formed at the expense of the albite. The general structure suggests that the albite is a later mineral 4 “ The Geology of Newton Abbot,” 1913, p. 60. 5 “ The Geology of Part of Carnarvonshire,” Geol. Mag., 1917, p. 18. 6 Proc. Roy. Soc. S. Aus., 1919. r. 328 also PL xxxi, fig. 2. UNUSUAL TO URM ALINE- ALBITE ROCK FROM ENOGGERA. 51 than the tourmaline, and has occupied sinuous cavities or replaced veins in the tourmaline rock. The size of the felspar crystals is variable, but usually one individual occupies the whole width of the vein. In addition to the veins, however, smaller crystals of felspar occur in the heart of the matted tourmaline. In other cases, particularly in the smaller veins and along portions of the margins of the larger, the felspar is present as groups of small crystals. (IV) CHEMICAL. In order to further investigate the nature of the minerals of the tourmaline- albite rock, quantitative chemical analyses were decided on. Part of one of the felspar bands was carefully removed, piece by piece, from the rock. This felspathic material was then ground very finely, and all traces of tour- maline removed by a powerful electro -magnet. In this way a small amount of the felspathic material was obtained in a fairly pure state. This was analysed by Mr. G. R. Patten in the laboratory of the Queensland Agricultural Chemist, with the result shown in the first column of Table 1. In order to provide the analyst with a larger quantity of material for the more difficult analysis of the tourmaline, a sample of the massive tourmaline from the edge of the granite was sent in preference to that which had been separated from the tourmaline- albite rock. The mineral submitted appears to be identical with that of the tourmaline- albite rock in colour, size and habit of crystals, and general appearance. The sample forwarded was seen to contain, however, as an impurity a considerable quantity of small red crystals of haematite. Consequently particular care was exercised by Mr. Patten in the determination of ferric and ferrous iron. The resulting analysis showed 7-43 per cent, of Fe203 out of a total of 99-95. This was assumed to be present in the form of haematite, and was consequently subtracted from the total, which was then recalculated to 100 per cent., with the result set out in the second column of Table 1. As a matter of interest, two further analyses appear in Table 1. These are analyses of the Brisbane Schist and of the Enoggera Granite (Pink Phase) respectively. Neither of 52 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. these analyses was made for the purpose of this paper, and neither is of rocks found in the immediate neighbourhood of the tourmaline- albite rock, so that no quantitative arguments could be safely based upon them. Of these analyses, No. 3 is that of a typical example of Brisbane Schist from the city of Brisbane. The Brisbane Schists are known to be made up of various types of altered rocks, but by far the greater part is micaceous schist, of which the analysis (3) should be quite representative. The Brisbane Schists in the locality under discussion are represented by very similar micaceous schists, so that the analyses quoted may give a general indication as to their chemical composition. Analysis No. 4 is that of a rock selected by the author as typical of the Pink Phase of the Enoggera Granite, and may be considered as a fairly safe guide to the chemical composition of the granite, which has all the mineralogical characteristics of the Pink Phase at this point. TABLE 1. — 1. 1 2. 1 3. 4. Si02 63-40 36-51 61-62 73-52 ai2o3 24-99 33-13 21-20 11-05 Ee„03 1-61 1-51 ; Nil EeO 0-28 10-30 1-93 3-15 MgO 0-37 5-23 1-77 1-03 CaO 0-42 1-58 1-59 1-70 Na,0 8-57 2-14 3-39 4-08 K,0 0-35 0-15 3-07 3-99 h;o+ 0-43 3-29 0-44 H00 - 3-03 0-16 Ti02 Nil 0-92 0-92 0-20 p2o5 0-18 0-17 0-15 MnO tr. 0-40 0-07 Br203 9-03 Total . . . . 100-12 100-00 100-56 99-48 Analysis 1 = Albite in Tourmaline -Albite rock. Analyst, G. R. Patten. Analysis 2 = Tourmaline in Tourmaline -Albite rock. Analyst, G. R. Patten. Analysis 3 = Brisbane Schist. Analyst, G. R. Patten. Analysis 4 = Enoggera Granite (Pink Phase). Analyst, G. R. Patten. An examination of Analysis 1 shows that the felspar present is undoubtedly albite. but a consideration of the UNUSUAL TOURMALINE-ALBITE ROCK FROM ENOGGERA. 53 relative proportions of soda, silica, and alumina shows that there is a considerable excess of the two latter. By using Harker’s or Osann’s tables we can easily calculate that the proportions of these three oxides in albite are very different from the ratios shown in the chemical analysis. Ratio in Albite. Na20. A1203. Si02. Actual 8-57 24-99 63-40 Calculated 8-57 14-15 49-74 Excess . . . . . . . . | 10-84 13-60 It is difficult to explain this very large excess, even after the other bases have been allowed their quotas of silica and alumina. The presence of andalusite and quartz suggested itself, but search for these minerals proved fruitless. Possibly many of the minute inclusions observed in the albite may be quartz and andalusite, but these could not nearly account for the large excess of silica and alumina. The conclusion was forced on one that the silica and alumina were present in solution in the albite in spite of the fact that Foote and Bradley investigated this very problem and came to the conclusion that “ no solid solution of quartz, corundum or nephelite in albite occurs which is greater than the apparent variation in composition due to the ordinary errors of analysis.”7 The excess of silica, anomalous though it seems, serves to explain in part another anomaly seen in the rock itself — viz., the absence of quartz in a tourmaline rock which (the structure suggests) has been formed by replacement. Turning now to the second analysis, we find it quite normal except in t e low percentage of water present. (V) DISCUSSION. (a) Mineralogical. Mineralogically the rock, consisting as it does almost entirely of tourmaline and albite, is almost unique. Only one 7 Foote and Bradley, “ Constant Composition of Albite,” Am. Jour. Sc., xxxvii, 1913, p. 47. 54 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. other rock showing this curious association has been described. Concerning this, Dr. J. S. Flett writes as follows : — Tour- maline-Albiie Rocks. — In two places within this sheet (347) rocks have been found belonging to a class which has not hitherto been described. They consist of tourmaline and albite, and are undoubtedly altered states of the killas or clay slates, seeing that they retain the banding and slip-cleavage which the killas alone displays.” The albite occurs as “ a colourless mineral in the clear bands of the rock.” Further, “ it forms grains of very small size and very irregular shape, which seldom show twinning of any kind. The tourmaline is in small prisms which have brown and bluish green colour.” 8 Dr. A. Harker, F.R.S., to whom a specimen of the Enoggera rock was sent, writes to the author concerning it, as follows : — “ The tourmaline-albite rock is interesting, and is apparently a rare type. The only other occurrence known to me is on the border of the St. Austell granite, in Cornwall, and has been briefly described by Flett in the Survey Memoir (Geology of Bodwin and St. Austell). I collected and sliced this Cornish rock some years ago, when Benson and I visited the district. It differs somewhat from yours, the albite being in little clear untwinned granules, easily mistaken for quartz, and the little prisms of tourmaline having a more pronounced parallel arrangement.” This communication from the eminent petrologist is especially important, since Dr. Harker alone has had the advantage of handling the specimens from both Cornwall and Queensland. The presence of tourmaline in localised patches (usually associated with quartz) in rocks intruded by a granite magma is, of course, quite a common phenomenon, and can in most cases be readily explained, but the presence of albite in such rocks is a matter which seems to admit of more than one explanation. The mineral may be original, it may be a product of metamorphism of pre-existing minerals, or it may have been introduced by the granite magma. With respect to the last possibility, the evidence is some- what conflicting ; for, although petrologists were, as the result of a great weight of evidence, forced to the conclusion that a transfer of soda from the invading to the invaded mass 8 Petrological Notes in “ The Geology of Bodmin and St. Austell/’ 1909, p. 103. UNUSUAL TOURMALINE- ALBITE ROCK FROM ENQGGERA. 55 has frequently taken place where the former is a basic rock, such transfer was not admitted in the case of granitic intru- sions.9 Allport, as early as 1876, in discussing the altered Cornish slates, pointed out that of the minerals of the granite 44 schorl, two varieties of mica, quartz, and felspar,” 44 felspar is the only one absent from the altered slates.”10 He was obviously unaware of the tourmaline-albite rock later described by Flett. In 1887 Rosenbusch,* 11 as a result of his work on contact phenomena in the Vosges, decided that nothing except a little boric acid had actually been added to the schists. In 1881 €r. W. Hawes,12 in describing the contact phenomena of the Albany granite, concluded that the addition of soda to the invaded strata 44 may be regarded as certain.” In 1886 Bonney13 pointed out that the differences between schists of regional metamorphism on the one hand and contact meta- morphism on the other lay 44 chiefly in the presence of felspar ” in the former. In 1894 Hutchings14 emphasized the fact that soda was transferred from basic intrusions but not from acid, but as the result of further work he came to the conclusion, in the fol- lowing year, that soda almost certainly was introduced from granitic masses.15 The albitites of Tilley16 and many pegmatites also force one to the conclusion that albite is a mineral which may be formed as the result of pneumatolytic, hydatogenetic, or deuteric processes, not only in basic but also in acid rocks, and as such may be expected in the metamorphosed rocks about a granitic intrusion. Flett assumes that the albite in the Cornish rock has not been introduced by the granite, for he writes : 44 The abundance of albite leads us to correlate these rocks with the felspar hornstones of the calc-flintas series, and they may represent 9 See especially Roth, Chemische Geologie, vol. iii, 1893. 10 Q.J.G.S., 1876, p. 408. 11 Die Steiger Schiefe und ihre Contact zone, 1877, p. 257. 12 Amer. Jrnl. Sci., 1881, p. 28. 13 Pres. Add. Q.J.G.S., 1886, p. 104. 14 Geol. Mag., 1894, p. 74. 15 Geol. Mag., 1895, p. 165. 16 Proc. Roy. Soc. S. Aus., 1919, p. 334. 56 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. tourmalinised states of these hornstones.” Further, he writes that the specimens “ show that albite is more resistant to pneumatolytic emanations than mica, as all the mica has disappeared, and this is in accordance with the frequency of albite in the china stones and kaolinised granites.”17 One cannot assume in the present problem the presence of bands of albite in the unaltered schists, felspars in this rock being conspicuous by their absence, and must conclude that the albite is the result of contact metamorphism, and has either been formed by the recrystallization of the minerals, already present in the rock or that it has been introduced from the granite magma just as the tourmaline was. The chemical evidence, such as it is, favours the former view, but it were safer to disregard this evidence altogether, for the analysis on which it is based is that of a sample of Brisbane Schist some miles distant. The field evidence and structure of the rock are decidedly in favour of the latter alternative ; and they suggest, further, that the tourmaline albite rock was formed in two stages, which, however, may have followed each other very closely. The first stage was the tourmalinisation of the micaceous part of the schist and the expulsion of the excess silica together with that of the quartz veins. This silica was subsequently redeposited in druses, such as are found in the neighbouring massive tourmaline rock which represents a part of the schist which was affected by this first stage only. The second stage saw the deposition of albite from hot aqueous solutions along the planes vacated by the quartz. (b) Structural. The writer has found a number of descriptions, some of them illustrated by plates and figures, of tourmaline contact rocks, with structures very closely resembling that of the Enoggera rock. In some of these cases the schistose structure seems to be independent of, and in other cases the direct result of, the intrusion of the granite. Dr. Bonney18 has urged that foliated rocks of the latter type should not be termed “ schists,” and agrees with Sir A. Geikie in the suggested revival of Dr. Boase’s old name “ cornubianite,” but has re- defined it as “ essentially consisting of quartz, mica, and 17 Op. cit., 1909, p. 103. 18 Op. cit. Note to p, 104. UNUSUAL TOURMALINE-ALBITE ROCK FROM ENOGK3EBA. 57 tourmaline.” Teall points out that this is “ approximately equivalent to the tourmaline hornfels of Continental petro- graphers.”19 He also states that “ the term hornfels is used by many writers as a general name for the innermost zone in regions of contact metamorphism, and thus applied even to schistose and banded rocks.”20 A classic example of tourmaline “ schists,” in which the schistosity is the direct result of contact metamorphism, is that described from Cornwall by Allport in 187 6, 21 where the “ original lamination of the fine sedimentary matter has been replaced by a distinct foliated texture.” He adds : <£ It should be stated, however, that a decided foliation is restricted to the immediate vicinity of the granite.” In describing the con- tact effects on the Mylor slates, Reid and Flett state: “ Close to the granite and within the zone into which granite veins extend the slates become much twisted, gnarled, and knotted, are often full of tourmaline, and the gaps between the twisted laminae have been filled up with streaks and lenticles of quartz of schorl rock or of chlorite.”22 These authors also describe the formation of “ muscovite- tourmaline schist with quartz from phy llites by heated solutions from the invading granite.” It seems that the effect of contact metamorphism on such rocks as slates and schists may be either to obliterate the schistosity or to greatly accentuate it, and both these types of alteration can be observed about the edge of the Enoggera granite. Rosenbusch has described a “ tourmaline hornfels. schistose in structure,” composed of tourmaline, staurolite, white mica, and quartz from the clay slates of the Vosges, but he points out that in contact rocks in this area the schis- tose structure is exceptional, and that as the innermost zone is approached all schistosity vanishes from the rocks and they become massive in character.23 Means, in describing tourmaline-bearing quartz veins from Ontario,24 speaks of “ narrow bands of highly altered country rock embedded in quartz,” and consisting of “a 19 British Petrography, 1888, p. 386. 20 British Petrography, 1888, p. 375. 21 Op. cit., PI. xxiii, fig. 6. 22 “ The Geology of the Land’s End District,” p. 9. 23 Quoted from Teall, op. cit., p. 375. 24 Economic Geology, 1914, p. 129. 58 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. highly tourmalinised schist ” made up of “ fine needles of tourmaline set in a matrix of fine crystalline muscovite.” These bands, beyond doubt, represent fragments of very highly altered country rock “ which was probably a diorite or a chlorite schist,” and which was “ first in part made schis- tose by dynamic metamorphism.” The rock which most nearly approaches in structural characters the Enoggera rock seems to be that described by Flett25 from Cornwall as “ tourmalinised slate (E 1759), Belovely Beacon,” a microphotograph of which is shown and which is described as ££ a secondary aggregate of quartz and schorl which has replaced the original slate,” and in which £* the cleavage and slip-cleavage of the slates are perfectly retained.” The writer has in an earlier section commented on the remarkable similarity in structure between the albite-tour- m aline rock of the Enoggera area and the unaltered Brisbane Schist. In 1914 he wrote: ££ The whole schist has been so beautifully replaced that the new rock under the closest scrutiny shows every characteristic of the normal schistose structure.”26 As the result of further investigation he sees no reason to modify that statement. The schistosity shewn in this interesting contact rock is undoubtedly older than and for the most part independent of the contact metamorphism. (VI) CONCLUSION. 1. The rock described is almost on the contact of the Enoggera Granite with the Brisbane Schist. 2. Mineralogically it is almost unique, consisting as it does almost entirely of tourmaline and albite. 3. Practically no quartz is present, but a chemical analysis of the albite shows remarkable excesses of both silica and alumina. 4. Structurally the rock is remarkably like the unaltered Brisbane Schists, the tourmaline of the altered rock corres- ponding in position and amount with the micaceous part of 25 Petrological Notes in “ The Geology of Bodmin and St. Austell,” 1909, p. 187. 2° Op. cit., 1914, p. 153. UNUSUAL TOURMALIN E-ALBITE ROCK FROM ENOGGERA. 59 the unaltered schist, while the felspar veins of the former correspond in number and type with the quartz veins of the latter. 5. In order to explain (I) the association of tourmaline and albite, (II) the absence of free quartz, and (III) the struc- tural nature of the rock, the following hypothesis is advanced : — The micaceous parts of the schist have been metasomatically replaced by tourmaline, and the excess silica, together with that of the quartz veins has been removed to be redeposited elsewhere. Following this alteration, albite was introduced from the granite and deposited among the veins vacated by the quartz and in any interstices in the matted tourmaline. 60 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. PLATE II. Figure L — Photograph of polished specimen of Tourmaline-Albite reck found replacing Brisbane Schist near its contact with th& Encggera Granite. Approximately natural size. Figure 2. — Microphotograph of same rock in ordinary transmitted lights X 18. Proc. "Roy. Soc. Q’land, Vol. XXXV. Plate II. NOTES ON ESSENTIAL OILS OF DAPHNANDRA AROMATIC A. 61 Notes on the Essential Oils of Daph- nandra aromatica. By T. G. H. Jones, B.Sc., A.I.C., and Frank Smith, B.Sc., F.I.C. (Read before the Royal Society of Queensland , 27th June, 1923.) Daphnandra aromatica (N.O. Monimiacege) is dis- tinguished from other of the genus by its aromatic bark and slightly fragrant leaves. Parcels of the leaves and bark, collected for us by courtesy of Mr. E. H. F. Swain, Director of Forests, were submitted to distillation in high-pressure steam, and the small amounts of oil collected have permitted of deter- mination of their general character. OIL OF THE BABK. One hundred and forty-five pounds of the dry bark yielded 200 cc. (-3 per cent.) of a dark amber oil, heavier than water, and possessing the characteristic odour of the bark, predominantly that of sassafras. The following constants were recorded: — a P07 7; [a]D— 1-9, MD20 P525. Acid, ester, and acetyl values, nil. Nothing was extracted from the oil by alkali bisul- phites, and a trace only of phenolic body by caustic alkali. When placed in a freezing mixture the oil became almost completely solid. Fractionation of 150 cc. of 130 mm. pressure gave 4 cc. at 70-80° C., 4 cc. at 80-128° C., and 140 cc. at 129-130° C. The last fraction, constituting the bulk of the oil, possessed the characteristic properties of almost pure safrol, Md25 10530, a\ f 1*096, [a]D — ‘5 ; and was confirmed as such by preparation of piperonylic acid, its oxidation product. Safrol, which constitutes about 95 per cent, of the bark oil of D. aromatica, is also, among the Monimiacete, the prin- cipal constituent of laurel leaf oil,1 and has been identified 1 Gildemeister and Hoffmann, “The Volatile Oils.” 62 PROCEEDINGS OE THE ROYAL SOCIETY OF QUEENSLAND. in the leaf oil of Atherosperma moschatum (Victorian sassafras).2 Its occurrence in the volatile oils of the Lauracese is common, notably in those of the Cinnamomums. The bark oil of C. Olivieri (Brisbane sassafras) contains 25 per cent.3 The lower fraction (4 cc.) was not sufficient in amount for identification of the optically active constituent, pro- bably a terpene the general characteristics of which indicated it to be laevo-rotatory pinene. THE OIL OF THE LEAF. The yield from the air-dried leaves was approximately •3 per cent. The essential oil was light greenish yellow in colour, and possessed an odour faintty resembling cinnamon. Its constants were — a\ f '9181 ; [a]D -j- 33*7 ; [n]D20 L489.. Ester value 20 ; acetyl value 50. Sodium bisulphite solution extracted 2-3 per cent, of a body which was apparently not cinnamic aldehyde; and caustic potash solution, *5 per cent, of a phenol giving a bluish green colour with ferric chloride. On fractionally distilling the amount of oil available (45 cc.) under 30 mm. pressure it was resolved into two principal fractions, viz.: — (1) Boiling at 70-80° C., and (2) at 145-154° C. The lower fraction possessed [w]D 1-478 and a + 54-2 in a 1-dcm. tube. Its odour and general characteristics indicated the presence of limonene, though the amount of the fraction was insufficient to permit of purification and identification. The behaviour on fractionation would show that the principal constituents of the leaf oil are a terpene (d-limonene ?) or terpenes, and a body, probably a sesquiterpene, comprising the bulk of the last fractionr [n]D 1-501, amounting to about 50 per cent, of the whole oil. It is hoped to further elucidate the composition of the leaf oil of D. aromatica when a larger quantity of material is available. -M. E. Scott, J.C.S. 101 (1912), 1612. 8 G. W. Hargreaves, J.C.S. 109 (1916), 751. CONTRIBUTIONS TO THE QUEENSLAND FLORA. 63 Contributions to the Queensland Flora. No. 2* By C. T. White, F.L.S., Government Botanist, and W. D. Francis, Assistant Government Botanist. (Text-figures 1-9.) (Read before the Royal Society of Queensland , 24th July , 1923.) Order CRUCIFERS. Cakile maritima Scop. Collected on the ocean beach, Stradbroke Island, Moreton Bay, by C. T. White. This herb, which is found in Europe, South Africa, and South America, is also common on the sea beaches of New South Wales. This is the first record of its occurrence in this State. Order PITTOSPORACE.E. Hymenospomm flavum F.v.M. Some specimens of this tree were recently received from Ravenshoe collected by Mr. J. B. Manuell and from the parish of Barron collected by Mr. D. Fraser, both localities on the Atherton tableland, North Queensland. The specimens differ from the more commonly known form in the under surface of the leaves, the peduncles and pedicels being more hairy. The fruits also are covered with a loose, floccose tomentum. The tree is common on the Eungella Range, North Queensland, and herbarium specimens from there are similar to Southern ones. It has also been recorded from Rockingham Bay by Mueller (Fragm. Phytogr. Austr. v. 210 and vi. 168). Order GUTTIFERiE. Calophyllum touriga sp. nov. Arbor magna ; ramulis glabris subteretibus ; foliis modice petiolatis (petiolis ca. 1-3 cm. longis) coriaceis glabris ellipticis prominente pellucido- punctatis (12-18 cm. longis, 5-8 cm. latis) ; inflorescentiis floribusque ignotis ; fructibus magnis (ad 10 cm. longis) ovoideo globosis obliquis apice acutis ; seminibus globosis (ad 2.5 cm. diam.). * No. 1 in vol. xxxiii., 1921, pp. 152-165. 64 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. A rain forest tree attaining over 100 ft. (30 m.) in height. Barrel not flanged at base, bark scaly. Apparently glabrous. A branchlet 4 in. (10 cm.) long measures 2 lines (4 mm.) in diameter at base. At intervals and frequently between the Calophyllum touriga sp. nov. A, leafy shoot. B, fruit. C, seed. All about half natural size. D, portion of leaf showing venation, X 3/2. insertions of the leaves the young branchlets are marked by transverse cicatrix-like rings. Leaves elliptical, obtuse or rounded or shortly acuminate at the apex, entire, coriaceous, margins recurved, lateral nerves close, parallel and almost tranverse, -5-1 mm. apart ; measurement of leaves 4J-7 in. (12-18 cm.) long, twice to three times as long as broad ; petioles about 4 in. (1-3 cm.) long, prominent^ grooved on the upper side. No flowers available. Fruit large, apparently subtended by five orbicular calyx lobes, exuding a milky juice when cut, globular- ovoid, oblique, apex pointed, slightly narrowed at the base, attaining 4 in. (10 cm.) in length, endocarp in dry or partially dry fruit crustaceous, about -5 mm. thick. Seed globular, about 1 in. (2-5 mm.) in diameter. (Text-figure 1.) Hab. : Bellenden-ker Range, at altitudes of 2,000-3,000 ft., C. T. White ; Boonjie, D. Fraser ; Gourka Pocket, Atherton Tableland, A. L. Merrotsy (fruiting specimens, January, 1923). Text -figure 1. CONTRIBUTIONS TO THE QUEENSLAND FLORA. 65 This tree, which is very plentiful in the localities enumerated, has a hark with a general resemblance to that of the Bolly Gum, Litsea reticulata. The timber is durable and is extensively used on farms for fencing posts and rails and for purposes in which hardwood is generally used. It also resembles the timber of Litsea reticulata, but is darker in colour and heavier in weight. Order MALVACEAE. Abutilon crispum G. Don. S.yst. i., 502. Collected at Bowen by Rev. N. Michael. This Indian and tropical and sub-tropical American plant was previously unrecorded for the State. Order STERCULI ACEiE . Helicteres Isora Linn. Collected on Settlement Creek, North-west Queensland, by L. Brass. An Asiatic plant only previously recorded in Australia from the Northern Territory. Order TILIACE^. Corchoms olitorius Linn. Collected on the Landsborough and Flinders Rivers of Central and North Queensland by C. T. White. A tropical Asiatic plant previously unrecorded for the State. Elaeocarpus eumundi Bail. The flowers of this species, which was named and described from fruiting specimens, have not been described previously. Following is a description of them : — Racemes secund, arising from the scars of fallen leaves sometimes at a distance of 10 in. (25 cm.) below the apex of the branchlets ; rhachis, pedicels, sepals, and petals silky pubescent ; rhachis about 2 in. (5 cm.) long ; pedicels slender, 4-6 lines (8-12 mm.) long ; flowers pale or greenish yellow ; sepals 5, narrow linear, acuminate, margins incurved, 4-5 lines (8-10 mm.) long ; petals 5, oblong, 5 lines (10 mm.) long, divided at the apex for about one-fourth of their length into about 20 linear lobes ; disc evenly and finely 10-toothed, glabrous ; stamens about 30, puberulent ; filaments about 1 line (2 mm.) long ; anthers narrow linear, about 2 lines (4 mm.) long, one of the two apical points recurved and exceeding 1 line (2 mm.) in length ; ovary ovoid, glabrous ; style slender, glabrous, 4-5 lines (8-10 mm.) long. Hab. : Flowering specimens collected Dec. 1922, at Fraser Island, by F. C. Epps. R.S. — F 66 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Order RUTACE^E. Boronia granitica Maid. & Betche, Proc. Linn Soc. N.S.W. vol. xxx., 357 (1905). This shrub, which was first found at Emmaville, N.S.W., has now been collected at Stanthorpe, Queensland, by Alex. Macpherson, W. R. Petrie, and E. W. Bick. Order MELIACEiE. Aglaia ferruginea sp. nov. Arbor ramulis juvenilibus dense rufo-pubescentibus ; foliis longe petiolatis impari- pinnatis, 2-jugis, petiolo communi et rhachide stellato-pube- scentibus ferrugineis, foliolis breviter petiolulatis submem - branaceis ellipticis vel oblanceolatis acuminatis, paniculis axillaribus ramulis stellato-pubescentibus ; floribus subsessili- bus glomeratis, calyce 5 lobo, lobis ovatis obtusis ; petalis 5 glabris imbricatis obovatis vel ovatis ; tubo glabro obconico crenulato ; antheris exsertis ; ovario ad basem puberulo. A small tree about 20 ft. in height with a barrel about 6 in. in diameter. Young shoots, branchlets, underside of leaves, and parts of inflorescence ferruginous pubescent with stellate hairs, the indumentum very dense on the young shoots, young branchlets, midrib, and primary nerves of the under side of the leaves. Leaflets 4 or 5, the lateral ones on petiolules of about 1 line (2 mm.) or less, the terminal ones on petiolules of about 4 lines (8 mm.) or less, thin or almost membranous, elliptical or oblanceolate, acuminate, the terminal ones nar- rowed at the base, margins recurved, lateral nerves numerous and parallel, 2J-5 in. (6-4-12-8) cm. long, twice to three times as long as broad. Panicles in the upper axils, mostly shorter than the leaves, the ultimate branches consisting of very dense globular, ovoid, or oblong clusters of flowers. Flowers crowded, subsessile, globular, about 1 line (2 mm.) in diameter. Calyx beset on the outside with brown stellate hairs, divided for about two -thirds its length into 5 obtuse, ovate lobes about 1 mm. long. Petals 5, glabrous, strongly imbricate, obovate or ovate, about 1-5 mm. long. Staminal tube glabrous, obconical, slightly exceeding 1 mm. in length, crenulate at apex, anthers slightly exserted. Ovary (perhaps rudimentary) globular, minute, pubescent at base. (Text-figure 2.) Hab. : Atherton Tableland, C. T. White (type) ; also received from the District Forest Officer, Atherton, without name of collector. This species is distinguished from its nearest Australian ally, Aglaia eloeagnoidea Benth., by the dense ferruginous hairs of its young shoots, branchlets, and under sides of leaves and by the dense globular, ovoid, or oblong clusters of flowers on the ultimate panicle branches. CONTRIBUTIONS TO THE QUEENSLAND FLORA. 67 Aglaia ferruginea sp. nov. A, flowering shoot about half natural size. B, portion of under surface of leaf, X 5/2. Order LEGUMINOS^. Templetonia Hookeri Benth. Specimens of this shrub have been collected at Settlement Creek, North-west Queens- land, by L. Brass, and constitute a definite locality record in the State for the species. The leaves in these specimens are occasionally 3-foliolate. 68 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Psoralea pustulata F.v.M.- The following are definite localities for this little-known species : Lower Settlement Creek, L. Brass ; Georgina River, E. J. Whelan ; Lawn Hill, E. Hann (all near Queensland-Northern Territory Border) ; Walsh River, Cape York Peninsula, T. Barclay Millar. Tephrosia coriacea Benth. This Northern Territory plant, previously unrecorded for Queensland, has been collected on sandstone ridges at Settlement Creek, North-west Queensland, by L. Brass. Atylosia cinerea F.v.M. This Northern Territory shrub, which was previously unrecorded for the State, has been collected at Branch Creek, North-west Queensland, and Settlement Creek, Queensland-Northern Territory Border, by L. Brass ; and at Townsville, North Queensland, by J. W. Fawcett. Rhynchosia acutifolia F.v.M. Definite Queensland localities for this species are — Gilbert River, F. v. Mueller ; Tate River, R. C. Burton ; near Railway Crossing, Tate River, C. T. White. R. C. Burton’s specimens bear pods containing 2-3 seeds, a peculiarity already noted by Bentham in Mueller’s specimens from the Gilbert River. Some specimens in the Queensland Herbarium simply marked “ Hann’s Expedition ” also have pods which are mostly 3-seeded. Dalbergia monosperma Dalz. This twining shrub, which is distributed in parts of India, China, Malaya, and Northern Australia, has been collected on the edge of Mangrove Swamps at Cairns by C. T. White, and was previously unrecorded for the State. Acacia lycopodifolia A. Cunn. This Northern Territory shrub was collected at Settlement Creek, North Queensland, near border of Northern Territory, by L. Brass, and is a new record for the State. It was recorded by Bailey in the 44 Queensland Flora,” ii., p. 483, as a Queensland species, and the locality Hammersley Range given. Hammersley Range, however, is in the north-west of Western Australia. Acacia flexifolia A. Cunn. A New South Wales species not previously recorded for this State which has been collected at Catfish Creek, Inglewood District, by C. J. Smith. Acacia myrtifolia Willd. has been collected in the following localities : — Logan River, Rev. B. Scortechini ; Russell Island, CONTRIBUTIONS TO THE QUEENSLAND FLORA. 69 Moreton Bay, Miss E. N. Parker ; summit of Glasshouse Moun- tains, F. M. Bailey ; Beerwah (associated with Acacia com- planata in Eucalyptus forest), C. T. White. In the “ Queensland Flora,” ii., p. 489, F. M. Bailey recorded A. amoena Wendl. from the Glasshouse Mountains. C. T. White recently collected a series of specimens of A. myrtifolia from that district, and on examination found them to be identical with F. M. Bailey’s specimens referred to A. amoena. A. amoena should, therefore, be deleted from the list of Queensland Acacias. J. H. Maiden, in ‘ ‘Forest Flora of New South Wales,” v., p. 185, had expressed his doubt as to this species occurring in Queensland. Acacia translucens A. Cunn. A Northern Australian species, not previously definitely recorded for Queensland, which has been collected at Massacre Inlet, Gulf of Carpen- taria, by L. Brass. J. H. Maiden, in his paper on the Tropical Acacias of Queensland, p. 24 (Proceedings of Royal Society of Queensland, vol. xxx.), excluded this species, as the only locality recorded in the 44 Queensland Flora” (p. 494) was “ Islands of the Gulf of Carpentaria.” The present record now definitely establishes the plant as a Queensland species. Acacia Muelleriana Maid. & Bak., Proc. Linn. Soc. N.S.W., viii. (2nd series), p. 515, pi. 25. A New South Wales species collected at Chinchilla by J. E. Young, and constituting a new record for the State. The Chinchilla specimens have slightly broader leaflets than those of the type, but resemble exactly specimens from Minore, N.S.W. (ex. Nat. Herb. Sydney). The plant is very different in appearance from any other Queens- land species of Acacia. Order MYRTACE^. Eugenia macrohila sp. nov. Arbor parva vel f rut ex diffusis ; foliis glabris, ellipticis vel oblanceolatis bre viter petiolatis obtuse acuminatis vel raro obtusis subcoriaceis ; floribus ignotis ; fructibus solitaris pedicellatis globosis puniceis ; seminibus solitariis. A straggling shrub or a small tree with an irregular or leaning barrel attaining about 6 in. diameter. Leaves elliptical or oblanceolate, narrowed at the base into a very short petiole, obtusely acuminate or less frequently obtuse at the apex, rather thick in texture, margins slightly recurved, at least in dried specimens, lateral nerves oblique, generally 70 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. 3 to 5 conspicuous on each side of the midrib, 1 \-2\ in. (3-2-G-4 cm.) long, 2 to 3 times as long as broad. No flowers available. Fruit solitary in the upper axils on slender pedicels attaining 8 lines or rarely reduced to 1 line in length, young fruit finely pubescent with minute appressed hairs, mature fruit attaining about \ in. (1-3 cm.) in diameter (when fresh), red, the pulp sweet and very pleasant to the taste. Persistent calyx-lobes 4, narrowly ovate or narrowly triangular, obtuse, 2-2 \ lines 'Eugenia macrohila sp. nov. A fruiting branchlet. B, mature fruit. <3, seed. D, leaf showing veining. All half natural size. (4-5 mm.) long. Endcarp thin, readily broken with the fingers. Seed solitary, globular or slightly ovoid, -J-J in. (6-8 mm.) in diameter ; hilum broad and nearly as long as the seed. (Text figure 3.) Hab. : Marmor, 26 miles south-east of Rockhampton, W. D. Francis, March, 1920. This species is allied to E. carissoides F.v.M., from which it is dis- tinguished by its narrow leaves, slender pedicels, and the elongated calyx -lobes crowning the fruit. CONTRIBUTIONS TO THE QUEENSLAND FLORA. 71 Eugenia Petriei sp. nov. Arbor, ramulis foliisque glabris ; foliis tenuiter coriaceis obovatis apice acuminatis basi in petiolum satis longum cuneatim coarctatis, costis lateralibus utrinque ultra 30 cum costa intramarginali a margine ca. 2 mm. remota conjugentes ; floribus mediocribus in cymam terminalem pauciflorum digestis ; calycis tubo turbinato (4 mm. longo), lobis 4, hyalino marginatis ; petalis 4, suborbicularibus, (ca. 2 mm. diam.) staminibus 4 mm. longis, stylo 1- 1-1*3 cm. longo ; bacco oblonga 1-sperma. Eugenia Petriei sp. nov. A, lowering shoot. C, fruit. D, flower bud. E, flower. F, petal. A and C about half natural size, D natural size, E and F X 2. A medium-sized tree, glabrous in all parts. Leaves thin- coriaceous, obovate, mostly 4-4| in. (10-11*3 cm.) long, and 1J-2 in. (3*8-5 cm ) broad ; finely veined with numerous close parallel veins, the main lateral veins If -2 lines (3-4 mm.) apart with finer veinlets between them , the intramarginal vein mostly about 1 line (2 mm.) from the margin, veins and veinlets conspicuous on both faces in the dried leaf, apex abruptly acuminate or more or less tapering into an acuminate point, base gradually tapering into a petiole of 2-3 lines (4-6 mm.). Mowers white, not very numerous, borne in terminal trichoto- mous cymes usually shorter than the leaves . Cymes pedunculate 72 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. or sometimes sessile ; peduncles, branches, and pedicels rather stout. Calyx turbinate, 2 lines (4 mm.) long, lobes 4, scarcely 1 line (1J mm.) long, margin hyaline. Petals narrowed at the base into a very short claw, suborbicular, about 1 line (2 mm.) diameter. Stamens very numerous, 2 lines (4 mm.) long.. Style slender, 5-6 lines (1- 1-1-3 cm.) long. Fruit blue, oblong,, about f-in. (1-3 cm.) long but not seen quite ripe, 1-seeded. (Text- figure 4.) Hab. : Fraser Island, W. it. Petrie. This tree in the past has been confused with E. cyanocarpa F.v.M., and when one of us (C. T. W.) was on Fraser Island in May 1921 Mr. Petrie drew his attention to it and E. cyanocarpa, which is also common there ; the two trees are quite distinct in the field, E. Petriei seems to be more or less confined to the creek banks, the branches usually overhanging the water. In botanical sequence this new species comes nearest to E. cyanocarpa the distinctions being as follows : — Leaves 2-4 in. long, f-1 J in. broad ; stamens \ in. long or more ; fruit globular — E. cyanocarpa. Leaves mostly 4-4J in. long, 1^-2 in. broad, stamens 2 lines long,, fruit oblong — E. Petriei. Order UMBELLIFERiE . Siebera Billardieri Benth. var. lanceolata. All the Queens- land specimens in the Queensland Herbarium represent the above variety. We have specimens from the following localities : — Macpherson Range, H. Tryon, C. T. White ; Helidon, F. M. Bailey ; Crow’s Nest, Kenny and White ; Cooroy, H. A. Longman ; Fraser Island, F. C. Epps, J. E. Young. Order COMPOSITE. Olearia oliganthema F.v.M. This New South Wales shrub, which was previously unrecorded for Queensland, was collected on the Macpherson Range, Killarney district, by C. P. Saunders. The Queensland specimens differ from the type in possessing larger flowers and in the involucral bracts being densely silky pubescent. The description in the “ Flora Australiensis ” was drawn up from meagre material, and the plant evidently shows a much greater variation than there described. Order EPACRIDEiE. Epacris longiflora Cav. Collected on Mount Lindesay by W. Hill. J. E. Young and C. T. White also collected it there recently, and noticed that it was common on cliff faces. We CONTRIBUTIONS TO THE QUEENSLAND FLORA. 73 also have specimens from Springbrook, Macpherson Range, collected by Mr. J. Tait. It was omitted from the “ Queensland Flora ” although recorded from Mount Lindesay in the “ Flora Australiensis.” This mountain is on the border of New South Wales and Queensland, but almost wholly within the latter State. The species is common in the sandstone flora of New South Wales. Order MYRSINACEiE . Ardisia bifaria sp. nov. Frutex erectus glabrus ; foliis alternis distichis subsessilibus pellucido-punctatis, oblongo- lanceolatis basi auriculatis ; floribus racemosis, racemis axillaribus 2-5 floris ; pedicellis capillaribus ; sepalis 4 lanceo- latis petalis 4 ovatis acuminatis ; staminibus 4 ; ovario ovoideo. Text-figure 5. Ardisia bifaria. sp. nov. A, flowering shoot; B, flowers, X 3. C, leaf showing venation. D, shoot showing attachment of leaves. A, C, and D, about half natural size. A shrub attaining 4 ft. in height, glabrous. Branches terete. Leaves alternate, dotted with pellucid glands, dis- tichous, subsessile, oblong-lanceolate, tapering but obtuse at the apex, cordate at the base, appearing oblique on account 74 PROCEEDINGS OF TPIE ROYAL SOCIETY OF QUEENSLAND. of the lower basal auricle often overlapping the upper surface of the branchlet, and the upper basal auricle overlapping the under surface of the branchlet, midrib immersed and a few indistinct primary veins visible on upper surface of leaf, midrib raised and primary veins slightly more evident on the under side ; measurement of leaf-blade, 1-2 in. (2-5-5 cm.) long, 4-6 times as long as broad. Flowers in short axillary racemes of 2-5 flowers, the rhachis about 1 line (2 mm.) long. Pedicels capillary, 3-4 lines (6-8 mm.) long, each subtended by a lanceo- late, concave bract about -|-line (1 mm.) long. Calyx divided to the base into 4 lanceolate sepals about f-line (1 mm.) long ; petals 4, very shortly united and overlapping at the base, ovate, acuminate, nearly line (3 mm.) long. Stamens 4, subsessile, lanceolate or narrowly traingular, subcordate, about 1 line (2 mm.) long ; slits of anther cells extending to the full length of the cells. Ovary ovoid, tapering into a fairly slender style. (Text-figure 5.) Hab. : Bellenden-Ker, common in lowland forests, C. T. White (flowering specimens, type) ; Glenallyn, Atherton Tableland, Gourka Pocket, Atherton Tableland, C. T. White. Order SAPOTACEJE. Lucuma castanospernram (C. T. White) n. comb. This species was described and named from fruiting specimens by C. T. White as Chry sophy llum castanosjpermum in Bot. Bull, xxi., p. 12, Department of Agriculture and Stock, Brisbane. Examination of flowers, recently collected on the Atherton Tableland by G. Curry, shows that the species should be placed in Lucuma. Following is a description of the flowers, which were not available previously Calyx-lobes 5, ovate, obtuse, imbricate, silky pubescent, about 3 lines (6 mm.) long. Corolla-tube shortly cylindrical or constricted about the middle, hirsute inside towards the base, about 5 lines (10 mm.) long ; lobes 5, their margins over- lapping towards the base, broadly ovate, obtuse, about 1 line (2 mm.) long. Staminodia 5, obtuse, alternating with the corolla-lobes and about two-thirds their length. Stamens alternating with the staminodia and inserted lower in the corolla-tube ; filaments slender, under J-line (1 mm.) long ; anthers ovate, almost 1 line (2 mm.) long. The presence of the staminodia in the flower and the absence of albumen in the seed indicate that this species should be assigned to Lucuma. CONTRIBUTIONS TO THE QUEENSLAND FLORA. 75 Sideroxylon Brownlessianum F.v.M. Fragm. vii., p. Ill ; F. M. Bailey, “ Queensland Flora,” iii., p. 959. Fruit, which were previously unknown, have been collected at Gadgarra, Atherton Tableland, by A. L. Merrotsy and C. T. White. Following is a description of them : — Fruit oval, 1 in. long and J-in. in diameter, 2-seeded. Seeds narrowly obovate, f-in. long ; testa dark brown, glossy ; hilum linear, about three-fourths the length of seed. Order APOCYNACE.E. Alstonia longissima F.v.M., Papuan Plants, 1 (v.), p. 91 (1877). Specimens of this plant collected at Port Moresby, Papua, by C. T. White, show that the species is identical with A. somers etensis , Bail., “ Queensland Agricultural Journal” i, pp. 229, 368 (1897). Bailey’s specimens were collected at Somerset, Cape York Peninsula, by F. L. Jardine, and his name must lapse in favour of Mueller’s earlier one. The species is figured on page 324 of Bailey’s “ Comprehensive Catalogue of Queensland Plants.” Order LAURACEiE. Cryptocarya foveolata* sp. nov. Arbor magna, ramulis juvenilibus pubescentibus ; foliis alternis petiolatis ovatis vel ellipticis obtuse acuminatus subtriplinervis prominente foveolatis utrinque reticulatis ; paniculis terminalibus vel axillaribus ; floribus breviter pedicellatis, perianthii tubo obconico, lobis ovatis ; bacca globosa. A tree attaining a height of about 100 ft. and a barrel diameter of about 2 ft. Barrel not prominently flanged. Bark brown, fairly smooth ; when cut, light brown, y^-in. thick on a dree with a barrel diameter of 1 ft. 9 in. Sap wood white. Young shoots pubescent. Leaves alternate, on petioles 2-4 lines (4-8 mm.) long, ovate, lanceolate or elliptical, mostly shortly and obtusely acuminate, a pair of lateral nerves prominent towards the base giving the leaf a triplinerved appearance ; the other lateral nerves are very few and distant, both surfaces reticulate. The most prominent features of the leaves are the one or two pairs of hollow glands (domatia) * Though this tree was originally named G. cinnamomifolia Benth. var. parvifolia by the late F. M. Bailey, the name parvifolia has already been applied specifically to a Philippine Island species of Cryptocarya By E. D. Merrill. 7 6 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Text-figure 6. Cryptocarya foveolata sp. nov. A, shoot from young tree. B shoot from a very large tree. C and E, fruit at different stages. I) leaf showing under side. All natural size. CONTRIBUTIONS TO THE QUEENSLAND FLORA. 77 situated along the midrib at the axils of the principal lateral nerves forming protuberances on the upper surface and with orifices on the under side. Measurement of leaf -blade 1-2J in. (2-5-6*5 cm.) long, 1 J to 3 times as long as broad. Panicles terminal and in the upper axils, very slender and raceme -like, generally shorter than the leaves. Flowers very shortly pedicel- late. Perianth turbinate or campanulate, about 2 lines (4 mm.) long, the obconical tube about as long as the lobes, pubescent inside ; lobes ovate, about 1 line (2 mm.) long. Stamens of the outer series nearly as long as the perianth segments ; anthers ovate, obtuse ; filaments broad, finely puberulent. Glands (6) at base of stamens, subsessile, ovoid or nearly globular, obtuse, puberulent at base, about -5 mm. in diametei. Three inner stamens lanceolate, obtuse, about as long as the outer ones, filaments broad and flat, finely ciliate on the margins. Three staminodia alternating with the three inner stamens on short broad stipites, broadly ovate, acute, cordate and puberulent at base, about 1 mm. long. Ovary immersed in the calyx-tube, narrowly ovate, puberulent with a few minute appressed hairs ; style cylindrical, glabrous, apparently articulate above the ovary. Fruit black, globular, about J-in. (1-2 cm.) in diameter, crowned by the small circular scar of the perianth tube. C. cinnamomi folia Benth. var. Bail. Cat. Queens. Woods 1886 and subsequent editions No. 313a ; C. cinnamomi folia Benth. var. parvifolia Bail. “ Queensland Flora,” pt iv., p. 1301 (1901). (Text-figure 6.) Hab. : Mount Mistake, F. M. Bailey ; Roberts Plateau, Macpherson Range, C. T. White ; Ranges eastward of Emu Vale, C. B. Saunders, W. D. Francis. This tree is allied to Cryptocarya cinnamomi folia Benth., from which it is distinguished by its smaller leaves containing large domatia and lacking a glaucous under -surface, and by its globular not-depressed fruit. Cryptocarya pleurosperma sp. nov. Arbor magna novellis sericeo pubescentibus ; foliis petiolatis alternis ellipticis obtuse acuminatis prominente triplinerviis supra nitidis ; floribus ignotis ; fructibus ovoideo-globosis ; seminibus rugoso- costatis, costis ca. 12. A tall tree, young buds silky pubescent. Young branchlets angular towards the growing points. Petioles somewhat flattened in the upper part or the leaf laminae shortly decurrent, 3-5 lines (6-10 mm.) long. Leaves alternate elliptical, shortly and obtusely acuminate, prominently triplinerved, transverse 78 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. veins visible but not prominent, upper surface smooth and glossy, under surface darker and duller, measurement of lamina 3J-5J in. (9-14 cm.) long, 1-5-2-5 times as long as broad. No flowers available. Fruit ovoid-globular, red, over 1J. in. Cryptocarya pleurosperma sp. nov. A, leafy shoot. B, fruit. Cy seed. D, seed in transverse section showing thick, ribbed putamen. All half natural size. (3-8 cm.) long. Seed ovoid-globular, rugose and very pro- minently longitudinally ribbed, ribs generally 12 in number, putamen walls 2-4 lines (4-5 mm.) thick. (Text-figure 7.) Hab. : Bellenden-Ker, C. T. White (type) ; Gadgarra, Atherton District, T. Fuller ; Johnstone Biver, Rev. N. Michael. In its triplinerved leaves and costate seeds this species resembles C. australis Benth., to which it is apparently closely allied. It is dis- tinguished from C. australis by its more strongly costate and larger very rugose seeds which are twice the size, and its broader and larger leaves. Order PROTEACEiE. Persoonia oxycoccoides Sieb. This species, previously known from Victoria and New South Wales, has been collected at Inglewood, west of Warwick, Southern Queensland, by C. J. Smith. CONTRIBUTIONS TO THE QUEENSLAND FLORA. 79 PLACOSPERMUM gen. nov. Flores hermaphroditi. Perianthium cylindraceum regularis vel parum irregularis, superne recurvum (segmentis 4, demum solutis ?). Stamina 4, antheris linearibus quorum 3 imper- fectis, filamentis infra medium perianthii insertis. Glandulse hypogynae 4, angustse. Ovarium subsessile, angustum, ovulis numerosis uniseriatis. Folliculus coriaceo-lignosus, pluris- permus. Semina compressa, samaroidea, late alata. Arbor. Folia alterna, integra, coriacea, pennivenia. Flores racemosi, racemis ad apices ramorum paniculatis. Flowers hermaphrodite. Perianth terete or cylindrical, recurved in the upper part ; segments 4, regular or nearly so. Anthers 4, but only one perfect in each flower, linear, with 2 linear parallel cells adnate to the connective ; filaments erect, linear or oblong, inserted below the middle of the perianth segments. Hypogynous scales 4, linear or setaceous, distinct. Ovary scarcely stipitate, narrow and terete ; ovules numerous ; style continuous with the ovary. Fruit a follicle with a thin, woody exocarp, 1 -celled. Seeds numerous, in a single row, flat and broadly winged ; testa membranous ; cotyledons thin and flattened. A tall tree. Leaves alternate, often crowded towards the ends of branchlets, entire and penninerved. Flowers singly pedicellate in racemes arranged in panicles at the ends of branchlets ; pedicels subtended by bracts. Derivation : From Greek plane, a flat body ; sperma, a seed ; alluding to the flattened seeds. This genus evidently constitutes a new tribe of the sub-family Grevilleoidese, as it cannot be included in any of the tribes described in Bentham and Hooker’s “ Genera Plantarum ” or Engler and Prantl’s “ Naturlichen Pflanzonfamilien.” The new tribe, for which we propose the name Placospeimese, could be placed between Embothriese and Banksise as under : — - Sub -family Grevilleoidece. — Flowers in pairs or single in the axils of the bracts. Ovary with 2 or numerous ovules. Fruit with 1 or numerous seeds. Grevillece. — Ovary mostly with 2, rarely 4, ovules. Embothrice. — Ovary with 4 or more ovules in two series. Placospermece. — Ovary with 15 or more ovules, in a single series. Seeds transversely arranged in the follicles. Banksice. — Ovary with 2 collateral ovules. Placospermum coriaceum sp. nov. Arbor glabra ; ramulis validis ; foliis petiolatis spathulatis vel oblanceolatis obtusis coriaceis, marginibus integris recurvulis ; paniculis terminalibus 80 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. foliis brevioribus ; bracteis concavis, ovatis vel triangularibus ; floribus pedicellatis, corolla superne recurva ; staminibus corolla fere sequantibus ; folliculis subglobosis obliquis, pluris- permis, seminibus ca. 20 uniseriatis, compressissimis late alatis. A tall tree, all parts glabrous. Branchlets thick, marked by the broad cicatrices of fallen leaves. Leaves spathulate or oblanceolate, obtuse, coriaceous, margins entire and minutely recurved, tapering into a petiole J-f in. (1-3-1-9 cm.) long ; midrib, lateral nerves and a few reticulate ones visible on both surfaces ; measurement of leaf-blade 4-7 in. (10-7-17-8 cm.) long 3-5 times as long as broad. Panicles shorter than the leaves. Bracts subtending the pedicels concave, ovate or triangular, over 1 line (2 mm.) long. Pedicels about J-in. (1-3 cm.) long. Perianth about |-in. (1-3 cm.) long. Stamens nearly as long as the perianth segments, the filaments of the fertle ones about 1 line (2 mm.) long. Hypogynous scales about 2 lines (4 mm.) long. Ovary terete, about 4 lines (8 mm.) long including the continuous style. Follicles subglobose, oblique, about 1J in. (3-8 cm.) in diameter. Seeds about 20 in each follicle, arranged in a single row and filling the follicle, flat and strongly com- pressed, broader than long, f-1 J in. (1-9 3-2 cm.) broad including the broad wing. (Text-figure 8.) Hab. : Mount Alexander, near Daintree, North Queensland, ex Queensland Herbarium, without collector’s name (flowering specimens, type); Reserve 418 Danbulla, Atherton district, Forest Ranger D. Fraser (fruiting specimens). To Mr. B. F. Kruger, Wood Technologist, Queensland Forest Service, we are indebted for the loan of a sample of the wood of the species. The sample is a cylindrical piece about 2^ in. in diameter, either a section from the barrel of a young tree, or from a branch. Examined in cross- section with a lens it shows no “ soft tissue” (wood parenchyma), a some- what unusual feature for a Proteaceous wood but which is also a charac- teristic of the wood of Embothrium Wickhami F.v.M. The vessels (“ pores”) are numerous, isolated, or in groups of two or four. The rays are fine and number from about 28-36 to the cm. in tranverse section ; in a tangential section they are seen to be from -5-1 mm. in height. The wood is light brown in colour. Order EUPHORBIACE^. Croton densivesfcitum sp. nov. Arbor ramulis dense stellato-pubescentibus ; foliis petiolatis, petiolis dense pubes- centibus ; foliis alternis raro suboppositis vel subverticillatis, ellipticis lanceolatis vel oblanceolatis acuminatis supra CONTRIBUTIONS TO THE QUEENSLAND FLORA. 81 Placospermum coriaceuvi gen. et sp. nov. B and C; follicles. D and E, seeds. F, reticulate surface of wings of seeds. G, a single flower. All half natural size except F (X^), and Gr (natural size). R.S. — G 82 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. glabratis subtus pubescentibus biglandulosis ; stipulis setaceis vel capillaribus ; racemis terminalibus dense pubescentibus, bracteis setaceis ; fioribus pedicellatis, calyce alte 5-lobato intus hirsutis ; capsulis dense stellato-pubescentibus, stigmatis bifidis ; seminibus subglobosis. Text-figure 9. Croton densivestitum sp. nov. A, flowering twig, about half natural size. B, fruit, about f natural size. C, terminal portion of raceme, X 3/2. D, seed, natural size. E, apex of leaf. F, stellate scales, X 10- Gr, base of leaf, under side, showing glands. X 4- A shrub, the branchlets inflorescence and under sides of leaves densely pubescent with stellate hairs. Branchlets terete. Petioles densely pubescent, J-l in. (-6-2-5 cm.) long. Leaves alternate, occasionally nearly opposite or whorled, elliptical, lanceolate or oblanceolate, shortly but not broadly rounded at the base, prominently acuminate, margins serrate or entire, upper surface nearly glabrous, midrib and primary lateral nerves visible on both surfaces, the reticulate ones obscure, glands situated on the under side of the leaf, one on each side of the base of the midrib where the basal lateral nerves join it, glabrous and obconical ; measurement of leaf blade 3J-5J in. (9-14 cm.) long, twice to thrice as long as broad. Stipules CONTRIBUTIONS TO THE QUEENSLAND FLORA. 83 when present setaceous or almost capillary, 1-2 lines (2-4 mm.) long. Racemes terminal, 2-3 in. (5-7*7 cm.) long, males in the upper, females in the lower portion. Bracts subtending the pedicels, setaceous, about J-line (1 mm.) long. Pedicels i-l line (1-2 mm.) long. Calyx hirsute inside, divided nearly to the base into 5 lanceolate lobes, about f-line (1*5 mm.) long. Petals (in male flowers) 5, lanceolate or oblong, obtuse, ciliate on the margins, f-line (1*5 mm.) long. Stamens 10-12, slightly shorter than the petals ; anthers ovate. Capsule stellate pubescent, tridymous, 3-4 lines (6-8 mm.) broad ; styles divided into 2 branches about 1 line (2 mm.) long. Seeds subglobose, pale brown, mottled or streaked with dark brown, about 2 lines (4 mm.) in diameter. (Text-figure 9.) Hab. : Harvey’s Creek, F. M. Bailey (Bellenden-Ker Expedition of 1889) ; (type) ; lowland rain forests, Bellenden-Ker, C. T. White, March 1922 (fruiting specimens). Among the Queensland species this one is apparently allied to C. arnhemica Muell. Arg. var. urencefolius Bail, (to which it was referred by Bailey in Report of Bellenden-Ker Expedition) and to C. acrony- chioides F.v.M. From the latter it is distinguished by its dense indumentum and long acuminate leaves, and from the former by its narrower almost non-reticulate leaves and smaller flowers and fruit. Order CONIFERS (TAXACTLE). Podocarpus spinulosa R. Br. This New South Wales undershrub, which was not previously recorded for the State, has been collected and noticed to be very abundant by C. T White on Stradbroke Island, in the sandy forest land between Dunwich and Point Lookout. The specimens (collected in March and July) bore leaves only. P. Ladei Bail. This species was described from specimens bearing leaves and immature fruit. A. L. Merrotsy has recently collected mature fruiting specimens from the same locality, Mt. Spurgeon, North Queensland, in wrhich the type material was collected. He described the trees as not exceeding 3 ft. in barrel diameter, and collected the fruiting specimens in Feb- ruary 1923. The mature fruit are bluish black, ellipsoid, fleshy, \\ in. (4 cm.) long, 1 in. (2.5 cm.) broad ; seed 1-1J in. (2*5-3 cm.) long, 8 lines (1*7 cm.) wide, the apex with a short, sharp, apiculate point. Bailey’s description and illustration are in the “ Queensland Agricultural Journal,” vol. xv. (1905), p. 899 pi. 22. 84 PROCEEDINGS OF THE ROYAL SOCIETYr OF QUEENSLAND. Order GRAMINEiE. Panicum ctenanthum F.v.M. This grass, which was previously unrecorded for Queensland, has been collected at the Gilbert River, North Queensland, by C. T. White. The specimens differ from the type in having most commonly 3 instead of only 2 spikes to the head, the spikes are also longer — 4 in. (10 cm.) long — but otherwise agree with the description of the type material. P. majusculum F.v.M. ex Bentham FI. Austr., vii., 482, 1878 ; Ewart and Davies, Flora North. Terr., 38, 1917. This grass, which was previously unrecorded for Queensland, has been collected at the Gilbert River, North Queensland, by C. T. White. The specimens differ from the original description quoted above in having the outer glume 5-7-nerved instead of 3 -nerved. Examination of type specimens kindly lent by Mr. Laidlaw from the National Herbarium, Melbourne, showed the outer glume to be often more or less distinctly 5-nerved. PHYSIOGRAPHY OF EASTERN NEW GUINEA, ETC. 85 Notes on the Physiography of Eastern New Guinea and Surrounding Island Groups. WITH SPECIAL REFERENCE TO THE VOLCANIC FEA- TURES OF THE RABAUL DISTRICT OF NEW BRITAIN. By C. H. Massey. (Plate III, and Four Text Figures.) (. Read before the Royal Society of Queensland , 24th July , 1923.) CONTENTS. I. — Introduction II. — General Physiography III. — History and Development of the Rabaul Volcanic System IV. — Present Conditions of the System V. — Conclusions — (i.) The Volcanic System (ii.) Physiography VI. — Description of the Plate . . 85 . . 87 . . 88 . . 90 . . 95 . . 101 . . 108 I.— INTRODUCTION. I have written this paper at the suggestion of those who think that the information given will be of interest and value to Australian Geology. By way of introduction, I must be allowed a few words of apology. I have had a certain amount of diffidence in preparing these notes, for several reasons. The whole area is so wide and so little known that there is not a great deal of reliable information available. The observations reported in this paper were made casually without any idea of future publication, and so lack the care and accuracy demanded in that case. A space of four years has elapsed since they were made, and though time allows maturer reflection, yet many facts are not so fresh to the mind as formerly. However, with thest limitations, I propose to discuss the physiographic features of Eastern New Guinea and the adjacent island groups, with special reference tc the volcanic features of the Rabaul 5^a/e . PHYSIOGRAPHY OF EASTERN NEW GUINEA, ETC. 87 area of New Britain. Most of the places referred to have been visited by the writer. For other information I am indebted to articles written by various officers of the garrison in the “Rabaul Record” (a publication issued monthly by the garrison at Rabaul for rather more than two years), and the excellent maps prepared by the Chief Surveyor, Rabaul, based on the Admiralty maps, corrected and brought up to date, from the latest information. I wish to thank Professor Richards, of the Queensland University, and his staff for much kindly help in the preparation of this paper. Other acknowledgments and references are made in the course of the paper. II.— GENERAL PHYSIOGRAPHY. The physiography of the area will be discussed in detail in later sections. It will only be necessary at the outset to draw attention to the more distinctive features. The general physiography of the area may be clearly seen in the sketch-map, text-fig. 1. The first thing that attracts notice is the parallelism in the trend-lines of Eastern New Guinea and New Ireland and the Solomons, and the apparent cross trend-line of the great peninsula and New Britain. In the arrangement of the mountain ranges of New Guinea something similar is seen. In the western part of the island the trend-line is west to east, the ranges extending nearly to the north-east coast. This trend seems to be continued in a parallel line in the great peninsula and New Britain. In the eastern part of New Guinea the trend-line is south-east. Parallel to the north-east coast- line is a series of volcanic islands, some active, rising directly out of the ocean upwards of 5,000 ft. They seem to be the peaks of a considerable range of mountains running parallel to the coast ranges and now submerged. This coast-line is generally bold throughout its length, and is flanked by series of coast ranges of varying altitudes up to 6,000 ft. The great peninsula is really a huge mountain massif built up of parallel ranges rising to a height of over 12,000 ft. in the Finisterre Range. The trend of these ranges is, as previously noted, west to east, crossing the trend of the coast-line and coast ranges. The general average height of the ranges of the peninsula seems to be from 5,000 to 8,000 ft. In New Britain, the average height, as far as is known, seems to be about 3,000 ft. with a few 88 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. peaks rising to 7,000 ft. New Britain, as far as is known? is entirely volcanic, and there are many craters both active and quiescent throughout its length. To the north of New Britain is a small group of islands known as the French Group. The main island is a great crater, partly submerged, known as Johann Albrecht. As observed inside,, from the deck of a steamer, the crater must have a diameter of three miles. The walls are perfectly perpendicular to a height of about 600 ft., while the water in the crater is. over 400 fathoms or 2,400 ft. deep. To the south of New Britain is one of the great deeps of the Pacific, known as. the Planet Beep. Depths up to 4,000 fathoms have been recorded. Fringing the east coast of New Ireland is a series of volcanic islands roughly parallel to the coast-line. St. John Island is noted for its mud-springs and fumaroles^ and contains at least one active geyser known as Balamus- son. These islands rise out of deep water over 1,000 fathoms. The Rabaul volcanic system is situated at the extreme eastern end of the island of New Britain, surround- ing the harbour known variously as Blanche Bay, Simpson- haven, or Rabaul Harbour. III.— HISTORY AND DEVELOPMENT OF THE It A BAUD VOLCANIC SYSTEM. The original crater of the Rabaul volcanic system is undoubtedly the depression known as Simpsonhaven. The definite walls surrounding the harbour, though now much weathered, composed of vast beds of yellow ash, pumice, and scoria, and the general shape, indicate this fact (text-fig. 2). Even in its original condition it must have been of considerable extent. At present the depression has a length of five miles and a breadth of two. The other craters are marginal to Simpsonhaven. The oldest seems to be Mount Mother, 2,500 ft. high, whose tall, steeply sloping sides indicate a typical tuff-cone (Plate III, fig. 1). This crater is quite extinct. Two further marginal craters were formed later, one at the foot of Mount Mother and the other on the South Daughter. These two craters appear much alike and no apparent distinction as to age is observ- able. They are both plugged, but contain active fumaroles. When Dampier visited Blanche Bay in 1699 he observed an active volcano on the north side of the bay. This probably refers to one of these craters. An eruption is reported to craters, the shore-line topography, and the probable direction of the local rift-lines. (89) 90 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. have occurred in a submarine crater, now known as Vulcan Island, on the western side of the harbour, somewhere about the year 1838. The latest crater is a cinder-cone formed on the margin of the South Daughter Crater. This was apparently formed and was violently active in the eruption of 1878. A full account of the eruption is given by the veteran Methodist missionary, Dr. Brown, in his ‘‘Reminiscences.’’ The following facts may be noted: — The eruption was preceded by very violent earthquakes accompanied by tidal waves which destroyed a large part of the shore-line. Clouds of steam arose from the water in a line from Matupit Island to Vulcan Island (text- fig. 2). When the submarine crater at Vulcan Island broke out, the steam-clouds across the harbour cleared. Finally the crater on the South Daughter (Cinder Cone) burst out with terrific force, throwing up material to a great height and forming a column of smoke upwards of 4,000 ft. high. These explosive eruptions continued for some weeks. There was no discharge of lava, but great blocks of pumice and rock (consolidated ash) with yellow and black ash were continuously thrown out. St. George’s Channel was a floating field of large blocks of pumice, and it is said no salt water was visible. The whole coast-line adjacent to the area was burnt out, and the waters of Blanche Bay and Simpsonhaven were at scalding heat throughout. All submarine life was destroyed. The successive eruptions of this system have built up vast beds of ash which dominate the local topography for a radius of many miles. There is no sign anywhere, as far as the writer could see, in a radius of twenty miles, of the original topography. Since 1878 there have been no further eruptions of this system, though many severe earthquakes have occurred. IV.— PRESENT CONDITIONS. The present conditions in the Rabaul district, as observed by the writer about four years ago, will be described under the following heads (i) Craters; (ii) Fumaroles; (iii) Submarine activity; (iv) Ashbeds; (v) Seismic features. PHYSIOGRAPHY OB' EASTERN NEW GUINEA, ETC. 91 (i) The Craters. — Four of the craters will be ■described — ( a ) The South Daughter Crater, (b) the Cinder Cone, (c) Matupit Island and Harbour, (d) Vulcan Island. (a) The older crater on the South Daughter (Plate III, fig. 2), known by the natives as Towurwur, is very similar to that at the foot of Mount Mother, and the follow- ing description will apply equally well to both : — The crater is breached on the harbour side by means of which there is an easy entry. The walls, which are steep, are built up of successive layers of yellow ash, pumice, and scoria, and in places covered with black cinder-ash from the cinder-cone. Around the margin of the floor may be seen masses of agglomerate, composed of blocks of hardened tuff and ash. The floor is covered with a smooth carpet of soft sulphur and ash mud. It is very hollow in sound and is probably not very thick. Both the floor and surrounding agglomerate are very hot. Numerous fumaroles are present around the margin. (&) The Cinder Cone (Plate III, fig. 3). — This crater is the latest in the system. It has the steep, evenly formed Mope of tuff-cones and is built up mainly of black cinder- ash. The crater itself is shallow and is filled up with ash and stones. In the centre is a small pit, probably the site uf the last stages of the 1878 eruption. There are numerous active fumaroles around the walls of the crater. (c) Matupit Island and Harbour ( see Fig. 2 and Plate III). — Below the craters previously described, on the western side and forming part of the harbour, is Matupit Haven and Island. The coaling station which was to form qoart of the German Pacific Naval Base is situated on the island in a well-protected position. There are many interesting features to be observed here. The island itself is built up mainly of ash, ash-mud, and sand. At its northern end, the island was joined to the mainland by a causeway until the 1916 earthquake, when it subsided considerably. The water enclosed by the island is very deep and the shore-line abrupt and steep. The island slopes away very gently to the south and becomes ultimately merely a large sandbank, slowly being built up by the action of currents and tides and effects of S.E. monsoons. Along its eastern side, at the foot of the South Daughter •craters, the water of the harbour is very hot and in places 92 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. gives off steam. All the evidence points to the fact that Matnpithaven is a submarine crater, but it is impossible to observe its relation to the other craters except that it appears to be associated with the rift system. (d) Vulcan Island (Fig. 2). — This is another sub- marine crater and it was active in the earlier stages of the 1878 eruption. At that time vast quantities of pumice and mud were thrown up, building the low island now known as Vulcan Island. On the edge is a small lagoon which is apparently part of the crater proper. The water is very hot, as is the adjacent mud. All these craters are explosive in type and there is no sign anywhere of lava-flows or dykes, except one very small outcrop of dioritic rock at the base of Mount Mother. There is only a very small exposure, and it is impossible to give any opinion as to its nature. (n) Fumaroles. — Along the western margin of the South Daughter craters and in the Cinder Cone, around the shores of Matnpithaven and at Vulcan Island, there are numerous fumaroles. They are continuously active. Some build up small sulphur cones and pipes, some are merely holes. The steam comes out in puffs at regular intervals. The steam is probably composed of water vapour and H2S and S02, as clusters of delicate crystals of yellow sulphur are deposited around the vents. The sulphur fumes arise in such a quantity that, when the S.E. winds blow, the atmosphere is distinctly disagreeable in Rabaul over two miles away, and even strong enough to discolour silverware. On closer examination, these vents are seen to follow two fairly definite lines — one in a north-and-south direction extending from the Cinder Cone to the old crater at the foot of Mt. Mother, and the other, commencing in Towur- wur, can be traced along a small gully in a south-west direction to the water's edge, and is in a direct line with Vulcan Island and passing through Matupithaven. This latter is in fact the old line along which steam arose imme diately preceding the 1878 eruption. Where these two lines intersect, there is quite a cluster of fumaroles and the rocks are very hot. Between the town of Babaul and Matupithaven, on a plantation formerly owned by the Hamburg Siidsee Aktien Gesellschaft, there is a series of. hot sulphur springs. PHYSIOGRAPHY OF EASTERN NEW GUINEA, ETC. 93 (hi) Submarine Activity. — In many places in the wicinity of the craters, the water of the harbour is very hot. The temperature of the water rises and falls rhythmically. This seems to indicate the presence of submarine fumaroles. Submarine eruptions of considerable intensity and in deep ocean water have been reported by the naval authorities off the south coast of New Britain. (iv) Ash-beds. — As mentioned previously, the whole of the Rabaul area is built up of ash-beds extending for a considerable distance south and west. Good sections may be seen in the deep and narrow ravines near Bita Paka, about fifteen miles south-west of Rabaul. Sections may also be seen in the road cuttings on the Namanula Road at Rabaul. The different eruptions and stages of eruptions can be clearly marked at Bita Paka in the varying features of the ash. Numerous beds of yellow and brown ash and pumice are interspersed by beds of black scoria. Occasion- ally, large lumps of consolidated ash, similar to the agglomerate at the craters, are found embedded in the ash. The observed thickness is at least about 400 ft. It is impossible to state the total depth, as the base level of the streams is still in ash. The writer regrets that circum- stances did not permit the making of a permanent record of the Bita Paka section. Judging by the presence of quantities of pumice and yellow ash, the original lava would seem to belong to the acid group. (v) Seismic Features.— The present activity of the area is further evidenced by continual tremors and occa- sional shocks of considerable violence. There has never been any accurate record kept, and needless to state there were no recording instruments. The area is not altogether suitable for recording instruments, on account of the soft nature of the ground. The facts recorded in this section are those observed by the writer and others who have been compelled to share many strange experiences of seismic energy. Distinction must be made between the different classes of shock — (1) The violent shocks of some amplitude affecting wide areas ; (2) The less intense shocks and tremors of medium amplitude affecting only local areas : 94 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. (3) Continuous tremors and microtremors, most of which are not felt. (1) In the first class must be placed violent earth- quakes such as shook the district in January, 1916. It was felt all over New Ireland and recorded in Sydney. A house was thrown over as far away as Toma to the west. It was accompanied by a tidal wave of considerable dimensions. A steamer of about 2,000 tons, lying at the wharf at Rabaul, was at one moment above the sheds and shortly afterwards was bumped on the bottom of the harbour. The wharf itself, which was specially built to withstand earthquakes,, was buckled up as if it had been crushed. The intensity and amplitude were so great that coconut palms were swayed to the ground and troops camped on Matupit Island became sea-sick. The narrow causeway which had until then joined Matupit Island to the mainland and formed a roadway sank several feet, cutting off communication except by boat. The stores in Rabaul suffered considerably. All their goods were thrown into confused heaps on the floor. Many of those who did not escape out of the bunga- lows quickly had exciting experiences with heavy articles of furniture which were thrown in all directions. Some wider tectonic origin must be sought for this class of shock. It is perhaps to be found in the readjusting and settling movements which must be frequently taking place in this somewhat unstable area. (2) The second class of shock consists of quakes of medium amplitude and less intensity than the former, and only affecting the vicinity of the craters. Some of these are sufficiently energetic to cause people to leave the bunga- lows or, if happening at night, to throw them out of bed. One of the more intense of these shocks occurred on Tuesday, 22nd May, 1917. It was felt at Kokopo, about twelve miles away, but not in the Duke of York Islands or New Ireland. It was preceded for some days by tremors and swarms of tremors with growing frequency and intensity. Immediately preceding the shock the movements ceased, at least visibly, for about thirty-six hours. This space of quietness was followed, about 9 a.m. on Tuesday, 2'2nd, by three violent shocks lasting upwards of a minute each. Each shock was preceded by loud rumblings, coming from the direction of the crater, becoming louder as the PHYSIOGRAPHY OP EASTERN NEW GUINEA, ETC. 95 shock approached. To those standing in the open, the ground seemed to roll in definite waves, which seemed to be travelling from the south-east — that is, from the direction of the craters. The bungalows rocked violently and threatened to fall from their foundations. Trees shook violently and posts of different lands moved visibly. Considerable damage was done to glassware and furniture. As far as could be judged without instruments, the shocks seemed to be travelling from the direction of the South Daughter craters. Fortunately, the intensity was not great though the amplitude seemed to be considerable. As no information is available as to the behaviour of the craters and fumaroles, and the interior condition being unknown, it is quite impossible to correlate this class of shock with any definite crateral condition, though all the evidence points to> the craters as the seat of origin. (3) Quite distinct from the former classes of shock are the frequent tremors and microtremors, most of which are not felt. The area seems to be in a state of continuous movement. The influence of these tremors, even when not felt, may be observed in any of the wells in the town of Rabaul. The water is in a continuously rippled condition. Y. — CONCLUSIONS. It is proposed to discuss the evidence outlined in the previous sections under two heads — (i) the volcanic system, (ii) the general physiography of the area. (i) The Volcanic System. — The ultimate origin of the Rabaul volcanic system is probably to be found in the great stress movements, to be described later, which folded and faulted the whole marginal area of the old continental mass that included Australia, New Guinea, and surrounding islands — Solomons, Hebrides, Fiji, Tonga, Kermadec, and New Zealand, and probably the East Indies. The volcanic systems of New Guinea and New Britain undoubtedly lie along zones of fracture caused by the earth movements. In this section, however, only the purely local conditions of the Rabaul volcanic system will be considered. The following subjects will be dealt with:— (1) The unstable conditions of the area; (2) Local rift system; Fig. 3. — Sketch-map showing the relation of Eastern New Britain and Southern New Ireland. PHYSIOGRAPHY OF EASTERN NEW GUINEA, ETC. 97 (3) Condition of the craters; (4) Causes of activity; (5) Age and future activity. (1) Unstable Conditions of the Rabaul Area. — The craters are situated on a very narrow peninsula, with the deep waters of St. George’s Channel and Blanche Bay on either side. The slope is steep. The eastern slope of Mt. Mother drops from a height of about 2,500 ft. directly into St. George’s Channel at its base. The other craters are similarly situated. A section line, commencing at Anir Island, drawn west and south through New Ireland, Mt. Mother, and South New Britain, gives a succession of high land and deep water (figs. 3 and 4). Operating in con- junction with these unstable topographic features are the periodic very high tides and the strains and stresses due to lunisolar earth warping. In addition to these local conditions of instability, attention may also be drawn to the association of the high mountain mass of New Guinea with the great ocean deep situated between that island and New Ireland. (2) Local Rift System. — As might be expected, there is evidence of a definite local rift system.. The fumaroles, as described in a previous section, have a definite linear arrangement in two directions, north-and-south and north- east-an d-sou th-west (fig. 2). At the point of intersection the activity is considerable. This assumption is supported by the fact that in the earlier stages of the last eruption great clouds of steam arose along the east-west rift-line across the deep water of the harbour. All the craters of the South Daughter group are arranged about the rift system, as well as Mt. Mother, Matupit, and Vulcan Islands (fig. 2). Should further eruptions take place, it seems likely that a new crater will be formed at the intersection of the rift-lines. There, as previously stated, quite a number of fumaroles occur in close conjunction, forming a natural outlet for any further explosive activity. (3) Condition of the Craters. — It is unfortunate that the craters are plugged, for it is impossible to observe their internal condition. Certain facts may, however, be deduced. The floor of the old South Daughter Crater rings very hollow and may be comparatively thin, covering a deep aid R.S. — H 98 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. definite vent. It is quite certain that there is a deep- seated nucleus of great heat. The floor itself is hot, and immediately under the surface of the mud too hot to touch. The surrounding fumaroles, the very hot rocks, and the fact that well-waters around Rabaul, two miles away, are inconveniently hot below 20 ft., all provide evidence of a centre of great heat. There is no direct evidence at the present time of a lava column of any kind. The yellow ash, scoria, and pumice of the 1878 eruption prove the presence of lava during that period. However, the continuous tremors and swarms, and the occasional greater tremors (coming from the direction of the craters), indicate the presence of stress movements there. If the observations made at the Halemaumau Seismometric Obser- vatory, Mt. Kilauea, Hawaii,1 provide any analogy, it is possible to assume the presence at least of a deep-seated lava nucleus. At Halemaumau these tremors and swarms have been definitely associated with recorded movements of the lava column. It is, however, possible that the tremors may be caused by other conditions present in the crateral area. (4) Causes of Activity. — All the evidence seems to indicate that the causes of continued activity are to be found in the combination of tensiona! stresses and strains and the rift system. The action of the strains and stresses on the rift system result in the maintenance of the heat of the central nucleus, mainly by pressure release. This heat is further developed by thermo-chemical reactions. Strains and stresses are set up by (a) the fact that the whole area is unstable and seems to be in a continual state of readjustment, (6) periodic high tides, and (c) luni- solar earth warping. All these factors operating in a rifted area will of necessity combine to set up crustal movements with accompanying earth tremors and shocks, and pressure release in the central heated nucleus. This seems to be the explanation of the present activity of the Rabaul volcanic system. The occasional outbursts of explosive energy may be due either to (a) the confinement of outflowing heat and 1 T. Jaggar : Seismometric Investigation of the Hawaiian Lava Column, 1920; Bulletin of the Seismological Society of America, vol. 10, No. 4. PHYSIOGRAPHY OF EASTERN NEW GUINEA, ETC. 99 gases, (b) sudden crustal warping of considerable extent, (c) the admission of large quantities of sea-water to the highly heated nucleus, or (d) a combination of one or more of these conditions. Normally, the fumaroles act as safety valves in allowing release of the gases up to a certain point. Considering the explosive nature of the eruptions of this system, it is probable that the rift system has played an important part in the later phases of activity, especially as the waters of the harbour are so closely associated with it. It seems probable, judging from the description, that in the 1878 eruption the east-west rift-line opened as the result of crustal movements, allowing the admission of vast quantities of sea-water to the intensely heated area below ; the sudden pressure release, rock fusion, gas expansion in the subterranean passages, oxidation, and other thermo- chemical mechanism providing the forces necessary to bring about a severe and continuous eruption. This is further borne out by the fact that the lava was thrown out entirely in the form of pumice and dust.2 It is also quite possible that this process is continuous in a small way, for it is practically certain that water percolates slowly into the heated interior. In addition to the waters of the harbour, there is the large catchment of water resulting from an annual rainfall of upwards of 100 inches. It is, then, quite probable that steam pressure and even the explosion of combinations of water vapour and other gases may be causes of some of the shocks, especially those accompanied by explosive rumblings. (5) Age and Future Activity. — Modern investigation in volcanic areas has added very considerably to our know- ledge of their history and formation. It is now generally considered “that explosive activity is an old-age feature of volcanism. ’ ’3 The same applies to yellow ash and cinder- cones which are associated with explosive activity. All the evidence goes to show that the craters of the Rabaul area represent the expiring stages of the activity of the system. The craters are more or less in a solfataric condition. There is now no evidence of anything but explosive action. The 2 See Prof. J. W. Judd: “ Krakatoa, ? > 1884. 3 Cf. Jaggar, loc. cit., in a full discussion of the principles of volcanism. c (100) Fig. 4. — Sketch-section across New Ireland and New Britain. PHYSIOGRAPHY OF EASTERN NEW GUINEA, ETC. 101 whole district is built up of vast beds of yellow ash and pumice, sprinkled with scoria. The latest crater is a cinder- cone. Although the activity of this system seems to be expiring, yet the conditions described in the preceding subsections indicate the presence of forces sufficient not merely to cause a continuance of quiet activity but a sudden renewal of an intense explosive character. (n) Physiography. — The following subjects will be dealt with under this subsection: — (1) General physiography — (а) The trend-lines of New Guinea; (б) New Britain an area of subsidence; (c) New Ireland an area of uplift : (2) Relation to the general topography of South- western Pacific : (3) Relation to the earth movements: (4) Relation to Australia. (1) New Guinea, New Britain, and New Ireland. — In discussing the physiography of this area, it must be kept in mind that we are considering the marginal portion of the ancient continental area which originally included Aus- tralia, New Guinea, and surrounding islands — Solomons, New Hebrides, Fiji, New Caledonia, Tonga, Kermaaec, New Zealand, and very possibly the East Indies.4 From its relation to the Pacific Ocean and the continental mass, the area has been subjected to great stress movements. The dominant trend-lines of this portion of the South- western Pacific are generally north-west to south-east. A difficulty is, however, presented by the general west-to-east trend of the great peninsula of New Guinea and the island of New Britain. Professor David, in the “Federal Hand- book,” describes New Britain as a probable virgation of the west-to-east trend of Western New Guinea. Professor Suess, in “Face of the Earth,” vol. iv, considers the direction of the volcanic range in New Britain to be a difficulty in his theory of the Australian arcs. The writer ventures to suggest a theory, based on many personal observations, that will account for the present topography of the area. 4 See paper by E. C. Abendanon: Jour, of Geol., vol. 27, 1919. 102 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. (a) Trend-lines of New Guinea. — An examination of the map (fig. 1) reveals some striking features. The moun- tain ranges of New Guinea seem to be arranged in two definite series grouped around a central massif which combines the trend-lines of both. One series has a definite west-to-east trend and the other a definite north-west-to- south-east trend. The former would include the main divide commencing at the western end of the island and extending eastward to the central group and continued almost to the north-east coast. An extension of this range, or at least parallel to it, are the mountains of the Finis- terre group in the great peninsula and the volcanic range of New Britain. The general direction of the Sepik River lends support to the natural grouping of the mountains. The second series would include the main divide extending from the central group south-eastward to East Cape and continued in the Louisiade Islands. Parallel to this series are the north-east coast ranges extending from Dutch New Guinea to East Cape and the long group of volcanic islands fringing that coast-line. This arrangement of the moun- tains suggests two distinct trend-lines instead of one long arcuate ridge. The triangular arrangement of the central massive group is strongly suggestive of accommodation to two distinct movements. ( b ) New Britain a Subsidence Area. — The volcanic islands fringing the north-east coast of New Guinea are undoubtedly the peaks of a considerable range of mountains now submerged. The general elevation of New Britain is much lower than that of either of the adjacent peninsula of New Guinea or of Southern New Ireland. The great crater of Johann Ahlbrecht, in the French group, is par- tially submerged, and has a depth of 2,400 ft. below the level of the ocean. The old Simpsonliaven crater at Rabaul is also partially submerged, and cannot represent its original elevation. Again, it is not probable that the present ash-beds represent its whole history. The earlier records have disappeared, presumably, beneath the ocean. Imme- diately to the south of New Britain is the Planet Deep. Soundings have indicated a widespread depth of upwards of 3,000 fathoms, with sinks of 4,000 fathoms deep and over. All this evidence and the general appearance of the land strongly suggested to the writer that New Britain is part PHYSIOGRAPHY OF EASTERN NEW GUINEA, ETC. 1 03 of an area of subsidence. Professor David, “Federal Handbook,” page 321, mentions “a sheer cliff of quartz schist” facing the north-east at an elevation of 8,000 ft. on Mt. Suckling in South-east New Guinea. This indication of downthrow on the western side of the area strongly supports this suggestion. (c) New Ireland an Area of Uplift. — Somewhere in the southern part of New Ireland there is a core of older rocks of a continental nature.5 The exact locality is unknown, as the rock specimens were apparently picked up in the stream-beds. But, as far as I am aware and from many inquiries made, New Ireland seems to be mainly coralline in formation. At the northern end, a few miles south of Kawieng, the rock outcrops are recent coral for- mation. Those who have crossed the range between Nama- tanai and the west coast, in the centre of the island, report a similar structure at considerable elevation. The northern extremity of the island is low in elevation, and the land gradually rises in a. southerly direction. At the southern end there is a huge piled-up mountain mass upwards of 7,000 ft., much higher than the rest of the island. Here the mountains fall steeply into the sea and have the appearance of a fault escarpment. The island groups off the east coast probably indicate a faulting and downthrow on the eastern side of the island similar to that reported by Mawson0 in New Hebrides, subsequent elevation being aided by volcanic action. To the writer the evidence all points to the fact that New Ireland is an uplifted area. The extreme elevation of the southern end is probably due to the fact that it was, previous to these movements, a land area. (2) Relation to the Topography of the South-iv ester n Pacific. — It may reasonably be asked, how does this theory fit in with the prevailing topography of the South-western Pacific ? For detailed discussions of the different parts of this area the reader is referred to the following works : — Mawson, “The Geology of New Hebrides,” Proc. Linn. Soc. N.S.W., 1905; Woolnough, “Geology of Fiji,” Proc. Linn. Soc. N.S.W., 1907; Professor Marshall, “Proceedings of GProc. Roy. Soc. N.S.W., 1882, vol. 16, pp. 47-51; Proc. Linn. Soc. N.S.W., 1905, p. 400. 104 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Australian Association for the Advancement of Science,” 1911, and the same author’s “Geology of New Zealand”; Guppy, “The Solomon Islands,” 1887; Suess, “The Face of the Earth,” vol. iv; Dana, “Manual of Geology,” 4th ed., 1895, p. 37; Prof. David, “Federal Handbook,” pp. 316-325. From a study of these works, certain facts emerge — (a) The linear arrangement of the island groups in a general north-west-to-south-east direction; (5) the prevail- ing north-west-and-south-east strike of folds where noted ; (c) the almost universal uplift of Tertiary limestones and coral reefs. These facts show that there is tectonic relation- ship between these groups. Mawson states that the South-west Pacific island groups are lined along great fold-chains concentric on the Australian foreland. The trend-lines of these island groups are practically continuous with those of Eastern New Guinea and New Ireland. The following ridges may be noted: — (a) New Ireland, Solomons, and New Hebrides; (5) Eastern New Guinea, Louisiades, New Caledonia, Norfolk Island, and the Auck- land Peninsula of New Zealand; (c) another definite ridge having the same general trend-line west of the second ridge between New Zealand and Norfolk Island and the Thompson Deep. The Fiji group represents either a curve to the east in the direction of the first ridge or a parallel ridge. New Caledonia probably occupies a smaller ridge parallel to (5). This grouping agrees with Professor Dana’s arrange- ment of his Australian Chain, and seems, in view of all the facts, to be the natural one. Prof. Suess, 1 1 Face of the Earth,” vol. iv, p. 301, makes the two branches of his first Australian arc link in Norfolk Island. This arrangement ignores the Gazelle Basin, an ocean deep, upwards of 3,000 fathoms. Again, the contour of the ocean floor shows a succession of great ocean deeps extending from the Aldrich Deep to north-east of New Zealand, through the Gazelle Basip, between New Caledonia and the New Hebrides, and the Planet Deep between New Guinea and the Solomons to t)ie coast of New Britain. These deeps have roughly the saifie direction as the trend of the land areas, and are suggestive of a long, somewhat irregular, trough in which New Britain is situated, lying between the two ridges first described. PHYSIOGRAPHY OF EASTERN NEW GUINEA, ETC. 105 These facts show that the theory seems to fit in quite naturally with what is known of the topography of the area. (3) Relation to the Earth Movements. — From what is known of the geology of the South-western Pacific and the further observations of this paper, it would appear that there have been three dominant series of earth movements in this area since Palaeozoic time — - (a) Resulting in the Tonga, Kermadec, and New Zealand ridge having a nortli-north-east trend. ( b ) Resulting in the Himalayan-Burmese-East- Indian trend-line, having in the East Indies and New Guinea a west-to-east direction and including New Britain. (c) A series of movements coming from the north- east creating the strong north-west-to-south- east trend-lines of the South-western Pacific. The ancient continental area, whatever its exact extent may have been, seems to have existed more or less in this area until some time towards the close of the Jurassic Period. It was then that the first of these series of move- ments took place, uplifting the marginal area and forming the Tonga-Kermadec-New-Zealand ridge. These move- ments must have depressed much of the surrounding land, enlarging and deepening the great depression known as the Thompson Deep. All the vast series of Jurassic, Triassic, and older rocks of New Zealand have been folded and elevated. Many minor movements of uplift and subsidence have taken place since, but the direction of the axis of the land has remained the same.7 Early in the Tertiary Age, the second series of move- ments commenced, extending into late Pliocene times.8 These were associated with the formation of the Burmese- East-Indian trend-line and building up the west-to-east ridges in New Guinea and New Britain. These movements were undoubtedly accompanied by strong volcanic activity, and the original craters of the Rabaul system seem to be at least as old as this period. Following closely on the second series of movements in the late Pliocene or early Pleistocene time, the third 7 Marshall: “Geology of New Zealand. ’ ; 8 Prof essor David: “Federal Handbook, ” p. 287. 106 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. series of movements took place. It resulted in the definite ridges and troughs between Australia and Fiji, forming the strong north-west-to-south-east trend -lines. This series of movements fractured the older west-to-east trend-line along the north-east coast of New Guinea. The volcanic islands of that coast and New Britain were folded or faulted downwards and the great ocean deeps created. The old schistose peniplain of Eastern New Guinea, New Ireland, and other groups of the area were elevated to a differential extent. Since that time there have been minor movements of adjustment complementary to the uplift when e.g. Torres Strait rift valley was formed, and perhaps the fault scarp mentioned by Professor David as occurring on It. Suckling. With reference to these movements, it is of interest to note that the trend-lines of the first and third of these series movements meet in the north island of New Zealand, repre- sented by the Tonga-Kermadec-New-Zealand ridge and the Eastern-New-Guinea-New-Caledonia-Norfolk-Island - and- Auckland-Peninsula ridge. The centre of the volcanic activity of New Zealand is found where these two lines meet. Active and extinct craters can be traced in both directions. Professor Bonney (“Volcanoes,” 1902, p. 260) states that the “north island (of New Zealand) is probably situated at the junction of two zones of weakness.” New Zealand geologists record a great outbreak of volcanic activity extending from the Miocene through the Pliocene period. (4) Relation to Australia. — The relation of these move- ments to the general uplift of Eastern Australia in late Tertiary times is not altogether clear. The following facts are interesting and instructive : — The coast-line of North-eastern Australia in its general trend is apparently related to the third series of movements. Dr. Walkom, in his paper on the Mesozoic of Eastern Australia, records a N. 30 W. trend in the folds of the rocks of that period. He merely states that they are post Lower Cretaceous. The dominant faults of North-eastern Aus- tralia have a north and north-western trend, this faulting being accompanied by the foundering of portion of the east coast. The great outbreak of alkaline eruptions in Aus- tralia is generally considered to be of Pliocene age. The PHYSIOGRAPHY OP EASTERN NEW GUINEA, ETC. 107 •evidence for the late date of the uplift of Eastern Australia is summed up by Professor David in the 4 ‘Federal Hand- book,” pp. 252, 287. Mention is here made only of the -obviously recent age of the canyons of the rivers of Eastern Australia. It is very probable then that, as far as Eastern Aus- tralia is concerned, the warping of the great peniplain was part of the third series of movements. I have ventured to suggest this theory as a possible explanation of the direction of the volcanic range of New Britain, and also as the result of personal observations made in Eastern New Guinea, New Britain, New Ireland, and the Northern Solomons. It is based almost entirely on physio- graphic evidence, and awaits confirmation or otherwise until a more detailed examination of the area is available. 108 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. VI. — DESCKIPTION OF PLATE III. Figure 1. — A photograph of Mount Mother, 2,500 ft. high, rising above the remnants of the wall of the old Simpsonhaven Crater. It shows the steeply sloping sides of a typical tuff-cone. Figure 2. — A photograph taken on the side of the Cinder Cone looking west, showing the old South Daughter Crater and its floor. In the middle distance may be noticed the northerly portion of Matupit Island and the isolated rock in the harbour known as Dawaria. The background is the western shore of the harbour, and shows the more or less even skyline of the old wall of Simpsonhaven Crater. On the side of the Cinder Cone, ash and lapilli, thrown up in the last eruption, may be seen. Figure 3. — A photograph of the interior of the crater of the Cinder Cone, showing the materials of which it is built and the small pit in the centre. Over the edge of the crater, on the extreme right, may be seen the southern shore of Matupit Island. Proc. Eoy. Soc. Q’land, Yol. XXXV. Plate III. Pig. 1. Fig. 2. Face page 108. Fig. 3. VOLCANIC ACTIVITY IN SOUTHERN QUEENSLAND. 109 Permo-Carboniferous Volcanic Activity in Southern Queensland. By Professor H. C. Richards, D.Sc., and W. H. Bryan, M.Sc., University of Queensland. (Plate IV., and Text-figure 1.) {Read before the Royal Society of Queensland, 24th September, 1923.) (I.) Introduction. (II.) The Silverwood-Luckv Valley Area, (III.) Petrology. (IV.) Comparison with other Permo-Carboniferous Volcanics of Australia. (V.) Earth Movements and Igneous Activity. (I.) INTRODUCTION. During the course of field investigations carried out in recent years in the Silverwood-Lucky Valley area, some 10 miles to the south of Warwick, the authors were particularly impressed by the fine development of great series of lavas and tuffs met with in several parts of the district. Both the flows and the associated pyroclastic rocks are for the most part of an acid nature, but more basic volcanics are represented in the upper part of the series. On account of their effective resistance to weathering, the rocks under discussion are generally the dominating topographical forms in their immediate locality, and in some cases cause important divides. The thickness of this great series of volcanics as exposed in the Eight Mile Fault Block is about 5,000 feet, but there is reason to believe that a large strike fault has caused duplication, so that a minimum thickness is in the neighbourhood of 2,500 feet. However, since the uppermost beds in this faulted area may not represent the last phases of vulcanicitv, and as the volcanic material found in the 110 PROCEEDINGS OP THE ROYAL SOCIETY OF QUEENSLAND. different fault blocks may not represent identical horizonsr the total thickness of Permo-Carboniferous volcanics which have been extruded in the area may far exceed this amount. The interest in this series was in part petrological, for many very beautiful examples of spheroidal and fiuidal structures have been observed, but this intrinsic interest was surpassed when sufficient evidence had been accumu- lated to definitely indicate that the series was Permo- Carboniferous in age, and interbedded with fossiliferous Upper Marine strata. The only other area where volcanics of this age are at all extensively developed in Australia, so far as the authors are aware, is in the Southern Coalfield area of New South Wales, and there the rocks are in marked contrast chemically with those under discussion. In the Drake area two series of lavas and tuffs have been described, both of which resemble in some respects those of the Silverwood- Lucky Valley area, but they are regarded by Andrew's1 as of Lower Marine, Permo-Carboniferous age. (II.) THE SILVERWOOD-LIJCKY VALLEY AREA. Faulted down into the Sdverw^ood Series of Andesitic Tuffs and Radiolarian Cherts of Devonian age, there are four blocks of different sizes, composed of Permo- Carboniferous material. Three of these are largely made- up of volcanic rocks, while the fourth contains much tuffaceous matter, but little in the way of lava flows. These blocks have been termed by the authors — (i.) Eight Mile Block, (ii.) Tunnel Block. (iii.) Stanthorpe Road Block, (iv.) Condamine Block. The first two have had a pronounced effect upon the topography, as there is a marked parallelism between the strike of the rhyolite and dacite flows and the backbone of the ridges. Apart from the areas of hornfels around the margins of the intruding granite of late Permian times, these areas of Permo-Carboniferous lava flows mark the 1 Report on Drake . . . N.S.W. Geol. Sur. Min. Res., No. 12., p. 9. VOLCANIC ACTIVITY IN SOUTHERN QUEENSLAND. Ill highest country. In the third block there is little in the way of lava flows, and there is no apparent effect upon the topography of the area, but in the fourth block the highest ground in the neighbourhood is marked by the interbedded rhyolite tuffs, and a very close relationship exists between the direction of the ridge and the strike of these tuffs. (i.) Eight Mile Block.— This is an area of eight scpiare miles, and is situated around and about the Eight Mile Range and the Rhyolite Range. The former range is so-called locally on account of its distance south of Warwick, while the latter, which runs N.N.W. from the Eight Mile Range, has been named by the authors on account of its petrological nature. This block is composed of mudstones, sandstones, grits, and conglomerates containing fossils belonging to the Upper Marine stage, and interbedded with them are mudstones containing Glossopteris and Noeggerathiopsis. Conformably bedded above the fossiliferous sediments there are rhyolites, dacites, andesites, and basalts, and with these are rhyolitic and dacitic tuffs. The rhyolites and dacites much preponderate over the andesites and basalts, while the sequence appears to have been from earliest to latest — rhyolite, dacite, andesite, and basalt. Interbedded with the rhyolite and dacite flows are large thicknesses of the tuffs. Approximately two-thirds of this block appears to be composed either of lava flows or of tuffs, so that we have between 5 and 6 square miles covered by their outcrops. (ii.) Tunnel Block. — This is a wedge-shaped block about 2 miles long by \ mile wide, and is situated north of the railway tunnel about 1J miles north-west of Cherry Gully Railway Station. The length, which is a little west of a north-south line, corresponds to the strike of the flows and tuffs. In its petrological character it much resembles the material in the Rhyolite Range further to the north, and it is composed almost entirely of rhyolite flows, some of which are porphyritic and others spherulitic and fluidal. Conformably bedded below the lava flows are mudstones containing Productus brachythcerus and shallow water sediments closely comparable with those below the volcanic material of the Eight Mile Block, so that the age of this 112 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. material is very similar to that of the Eight Mile Block. The relationship between the strike of the flows and the direction of the ridge is very marked. (in.) Stanthorpe Road Block. — This is about 10 miles south of Warwick, and is composed of sediments belonging to the Lower Marine stage, the Lower Freshwater stage, and the Upper Marine stage. Interbedded with the last-named are one or two thin flows of rhyolite and a certain amount of tuffaceous material, but the importance of this occurrence, compared with the other blocks, is very small. (iv.) Condamine Block. — This is situated about the Condamine River, 14 miles south-east of Warwick. Inter- bedded with the Upper Marine sediments is a series of rhyolite tuffs about 700 feet thick. Conformably bedded beneath the tuffs are tuffaceous mudstones containing abundant remains of Trachypora Wilkinsoni, &c., while above them are rhyolite grits containing Martiniopsis subradiata, Trachypora Wilkinsoni, Productus brachy- thcerus, &c., so that there is little doubt as to the Upper Marine age of the tuffs. General Nature of the Activity, c See. Rhyolitic and dacitic effusions much preponderate, and with the exception of the Eight Mile Block there is little other than Rhyolite flows and tuffs. The fossils associated with the effusions are those characteristic of either the Upper Marine Stage or the Upper Freshwater Stage, but all are of the shallow water type. Probably the lowest portions of the Eight Mile Block and the Tunnel Block material were laid down in a shallow sea, but the great bulk of the volcanic material appears to have been extruded under terrestial conditions. At the Condamine and Stanthorpe Road Blocks, however, all the tuffaceous material appears to have been laid down as shallow marine deposits. The rhyolitic and dacitic effusions were interrupted by extensive explosive phases, and in the Eight Mile Block the Eight Mile Range is composed almost entirely of dacite flows and tuffs, while the Rhyolite Range is made up of rhyolite flows and tuffs. Much of the rhyolite was very viscous, as it frequently shows well-developed fluxion structure. VOLCANIC ACTIVITY IN SOUTHERN QUEENSLAND. 113 The concluding effusions appear to have been dense andesites and basalts, and although there is not any tuffaceous material associated with these latter flows, there is, at one point on Rhyolite Range, a very coarse rhyolite breccia immediately underlying the andesitic material, indicating quite local and intense volcanic explosion previous to the andesitic effusions at that point. The volcanic activity was characterised by^ — (a) An increasing basicity as it progressed; (&) The amount extruded being inversely propor- tional to the basicity. (III.) PETROLOGY, As indicated above there are in the thickness of at least 2,500 feet of volcanic material, both flows and tuffs. The flows range from acid to basic in character, and the former much predominate. Many of the fine-grained rhyolite tuffs are difficult to distinguish from flows, and recourse to microslides is necessary. Except in the Eight Mile Range, which is largely clacitic, almost the whole of the volcanic material is rhyolitic in character. Rhyolites. These are variable, but both lithoidal and fluidal types are common, while spherulitic structure is frequently developed in a striking manner in the hand specimen. The colour varies through pink, lavender, and grey, to white — the last especially on the weathered surfaces. Porphyritic crystals of quartz, orthoclase, and plagioclase occur, and frequently spherulites have developed around the felspar crystals. The descriptions of the three following rhyolites may be regarded as representative of the various flows in the several blocks under consideration : — Rhyolite from Eight Mile Range, near the Summit, Portion 2161, Parish of Wildash. This is a deep lavender- coloured rhyolite containing numerous small phenocrysts of sanidine and very occasional small rounded crystals of quartz in a very fine groundmass. Fluxion structure is shown well in the hand specimen. The specific gravity is 2-628. Under the microscope its holocrystalline character is apparent, and one sees idiomorphic phenocrysts of r.s. — i 114 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND* orthoclase varying from 0-5 to 0-75 mm. in length set in a groundmass of a traehytic nature and composed of quartz, felspar, and small dark rods and granules of a mineral which may be dark-brown hornblende. The rock has been analysed and the results are shown on Table I. {Micro,. 359.) 2 Rhyolite from The Hump, Rhyolite Range, Portion SvT Parish of "Wildash. This rock, which is typical of much of the material of Rhyolite Range and of the Tunnel Block areas, is pale-pink on the fresh surface and white on the weathered surface. It shows fluxion structure very freely,, and here and there through the very fine groundmass there is an occasional crystal of felspar. Under the microscope a few crystals of cloudy orthoclase are seen, set in crypto- crystalline groundmass through which there are frequent streaks of secondary silica corresponding to the lines of flurion. Density 2-561. {Micro. 452.) Spherulitic Rhyolite from The ITump, Rhyolite Range. In the hand specimen there are isolated rounded spherulites:, of cloudy felspathic material set in a greenish-grey ground- mass, giving much evidence of secondary silicification. The spherulites average approximately 5 mm. in diameter, and occur frecpiently enough to be separated from one another by a few millimetres only. Under the microscope the groundmass shows as a cryptocrystalline mass through which there is much secondary silica. The spherulites. appear to be masses of kaolin stained brown by limonite„ Density 2-526. {Micro. 453.) Dacites. The dacites are difficult to distinguish from some of the rhyolites, but have a more melanocratic appearance. They are markedly porphyritic and are of two types, one showing quartz and felspar phenocrysts, and the other felspar phenocrysts only. The descriptions of the two following rocks will serve; to illustrate very well the main characteristics of the dacites. Dacite from Portion 1663, Parish of Wildash, Eight Mile Creek. In the hand specimen it is a dark-grey colour with abundant clear rounded crystals of quartz and stout crystals of a pink plagioclase. Under the microscope the 2 These numbers refer to the petrological collections in Dept, of Geology, Univ. of Qld. VOLCANIC ACTIVITY IN SOUTHERN QUEENSLAND. 115 phenocrysts give much evidence of having been attacked by the matrix, as the edges are rounded and sometimes the crystals are embayed. The felspar phenocrysts sometimes form composite masses up to 4 mm. in diameter, but their average length is about 2 mm. The felspar phenocrysts give an extinction such as to indicate medium an desine (Abco- — An35). The groundmass is hemicrystalline and hyalopilitic, with crystals of quartz and felspar, together with small rod-like crystals of a ferromagnesian mineral which was perhaps originally augite but is now chlorite. Density 2-650. {Micro. 370.) Dacite from the centre of Portion 2274, Parish of Wildash, Eight Mile Range. In the field this rock was difficult to distinguish from the first rhyolite described. It is deep lavender in colour, with small phenocrysts of pink felspar. Under the microscope the phenocrysts are seen to occur up to 1 mm. in length and to show multiple twinning. It is difficult to determine them, but they appear to be of the andesine-oligoelase type. The groundmass is pilotaxitic and contains abundant small allotriomorphic crystals of quartz and felspar set in amongst the cryptocrystalline material. Density 2-609. {Micro. 455.) Andesites. The andesites vary from light-grey to purple in colour and from markedly porphyritic to non-porphyritic types. The three following rocks will serve to illustrate the- important characteristics of the andesites : — Anclesite from the top of Rhyolite Range, due east of Bald Mountain. In the hand specimen this is a purple- brown in colour with abundant phenocrysts of plagioclase of various sizes but frequently equidimensional. Under the microscope the phenocrysts of plagioclase show marked twinning and are determined as medium andesine (Ab65 — An J|J , from the extinction angle. Some of the felspars show definite zoning. The groundmass is hyalopilitic and fluxion structure is very evident. Little other than minute crystallites of plagioclase and glassy material may be seen in the base. Abundant granules of ilmenite, partly weathered into leucoxene, occur through the field. Density 2-691. {Micro. 447.) 116 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Andesite from Portion 1663, Parish of Wildash, Eight Mile Creek. This is a light-grey slightly porphyritic type, which has a specific gravity of 2-747. Under the microscope small phenocrysts of plagioclase np to 1-25 mm. long are seen set in a hyalopilitic gronndmass. Fluxion structure is well developed, and in the groundmass there are granules of augite up to 0-25 mm. in diameter. These granules also occur as inclusions in the felspar phenocrysts. There is much alteration, calcite and epidote being abundant. This rock has been analysed, and the results may be seen in Table I. {Micro. 361.) Andesite from the east of Portion 1179, Parish of Wildash, Eight Mile Range. In the hand specimen the rock appears light-grey in colour and has abundant phenocrysts of milkv-white plagioclase and dark augite set in a grey base. Under the microscope the phenocrysts show much zoning in some cases, and those which are not zoned give an extinction angle indicating basic andesine. The plagioclase phenocrysts range up to 3 mm. in length and the augite crystals up to 1 mm. long. The latter are usually fresh and often show twinning. The phenocrysts of plagioclase are more frequently zoned and perhaps more basic than in some of the other andesites. This andesite, however, is representative of a great thickness of material. The groundmass is rather of the pilotaxitic type and distributed through it are abundant granules of magnetite. Density 2-639. {Micro. 394.) Basalts. These are not at all abundant, but they are always very dense, fine-grained, and rather platy in character. The following one, which has been analysed and whose analysis is given in Table I, may be taken as typical : — Basalt from centre of portion 2161, Parish of Wildash, Eight Mile Range. In the hand specimen it is very compact and heavy, having a density of 2-847. It is very fine grained and weathers to a deep-brownish soil. Under the microscope one notes its hemicrystalline nature with a hyalopilitic groundmass showing marked fluxion structure. Throughout the field there are occasional lath-sliaped crystals of medium andesine (Ab60 — An40) and stout crystals of augite. These are set in a mass of crystallites of plagioclase, augite, and magnetite. {Micro. 364.) VOLCANIC ACTIVITY IN SOUTHERN QUEENSLAND. 11 1 Breccias and Tuffs. These vary in size from quite coarse breccias, with fragments the size of walnuts down to very fine tuffs. Not uncommonly one finds interbedded layers of material of widely different coarseness. Some of the breccias have been much silicified and consist of angular fragments of cryptocrystalline rhyolitic material cemented together by secondary quartz and chalcedony. The breccias all appear to be rhyolitic in character, which is, perhaps what one would expect, while the tuffs are rhyolitic, dacitic, and andesitic in nature. Chemical Characters. The accompanying five analyses and their norms serve to indicate the general chemical characters of the volcanic rocks. The rhyolite and the basalt are both in a fresh condition, but the rhyolite tuff and the two andesites show by the carbon dioxide and water contents that they have undergone a considerable amount of alteration. The lavas are all very much richer in soda than potash, and their most marked character is their deficiency in potash without any corresponding increase in soda content. For example, the world’s average rhyolite has about 3J per cent, of soda and 4 per cent, of potash, so in comparison the rhyolite in question has a normal soda content but is deficient to the extent of 2 per cent, of potash. The world’s average basalt has about 3 per cent, soda and half that amount of potash, while the basalt here shows considerable deficiencies in both, more especially the potash. If one compares the lime and magnesia contents it is found that they are about normal. In comparison with the Cainozoic lavas of Southern Queensland, the Permo-Carboniferous rhyolite has higher alumina, lower lime, slightly higher soda, and much lower potash. The Permo-Carboniferous basalt, in comparison with the Cainozoic basalts, has higher alumina, lower total iron oxides, higher lime, and lower alkalies. A comparison with the analyses of the volcanic rocks of the Southern Coalfield of New South Wales is made later in the paper, 118 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Table I. — Chemical Analysis of Volcanic Rocks. — Rhyolite . Summit, Eight Mile Range, Portion 2161. Rhyolite Tuff, Lucky Valley Creek, Portion 18v. Andesite, Portion 1663, Eight Mile Creek. Andesite, Portion 1663, Eight Mile Creek. Basalt, Portion 2161, Eight Mile Range. Analyst. G. R. Patten. G. R. Patten. G. R. Patten. G. R. Patten. G. R. Patten. Si02 73-93 73-87 55-49 54-25 53-34 ai203 . . 16-85 12-92 18-63 19.46 18-73 Fe203 . . 0-60 nil. 1-89 3-16 1-72 FeO MO 2-32 5-41 4-27 6-65 MgO 0-43 0-83 3-28 2-75 5-10 CaO 0-63 2-83 6-09 5-81 8-75 Na20 3-61 3-23 4-11 3-56 2-30 k2o 1-88 1-07 0-11 0-62 0-61 h2o+ .. 0.47 0-43 l 2-16 2-19 1-03 h2o- .. 0-15 0-09 0-17 0-22 0-21 C02 nil. 1-75 1 -19 1-97 nil. Ti02 0-45 0-32 1-70 1.82 1-80 P205 . . 0-07 0-03 0-14 0.21 0-15 MnO 0-02 0-04 0-09 0-10 0-17 Total 100-19 99.73 100-45 100-39 100-56 Spec. Grav. 2-628 2-652 2-747 2-753 2-847 Norms. Quartz . . 43-74 47-04 14-40 19-56 9-00 Orthoclase 11-12 6-67 0-56 3-34 3-34 Albite 30-39 27-25 34-58 29-87 19-39 Anorthite 2-50 1-95 21-96 15-85 38-92 Corundimi Diopside 8-06 5-61 3-67 7-14 2.94 Hypersthen e 1-89 5-80 13-88 9-14 19-32 Magnetite 0-93 2-78 4-64 2-55 Umenite 0-76 0-61 3-19 3-50 3-50 Apatite . . 0-34 0-34 0-34 0-34 0-34 Calcite . . 4-00 2-70 4-40 Water 0-62 0-52 2-33 2-41 1-24 Total 100-35 99-79 100-39 100-19 100-54 Classifi- 1.3.2. 4. 1.3.1. 4. II-4.3.5. II.4.3.5. II. 4(5). 4. 4 cation (IV.) COMPARISON WITH OTHER PERMO-CARBON- IFEROUS VOLCANICS OF AUSTRALIA. (a) General. — The volcanic series under discussion cannot be closely correlated with any similar series in Australia, and indeed this is one of the reasons for its VOLCANIC ACTIVITY IN SOUTHERN QUEENSLAND. 119 interest. However, it may be instructive to compare the Silverwood-Lucky Valley flows and tuffs with various other occurrences which have some features in common with them. It will be noted that where petrological comparisons reveal similarities, a consideration of the respective ages shows a lack of harmony, and on the other hand where the ages are in agreement the petrological characters are essentially different. (&) The Southern Coalfield of New South Wales. — Harper has described a series of lavas and tuffs about 1,000 feet thick from the Cambewarra-Kiama districts, the age of which agrees closely with the Silverwood-Lucky Valley Series. '‘These consist of seven submarine” . . . . (followed by) .... “two terrestrial lava flows, intercalated with beds of Permo-Carboniferous age consist- ing mainly of volcanic tuffs. ”3 Of these flows the ‘ ‘ Blow- hole Plow” “was apparently the first manifestation of volcanic activity in the shape of lava flows during the Upper Marine epoch, within the Southern Coalfield, and it is interbedded with tuffs containing marine fossils.”4 With regard to the uppermost “Minmurra Flow,” Harper states that “The horizon occupied is about 120 feet above the top of the Cambewarra flow or its equivalent boulder horizon, ’ ’5 6 which marks the upward limit of marine fossils. The Berkley Plow is also interbedded with the freshwater sediments overlying the Upper Marine. This evidence forces on one the conclusion that the voleanics of the Southern Coalfield were practically con- temporaneous with those of the Silverwood-Lucky Valley area, but here all resemblance ends. Chemically the two groups are essentially different, and this difference is reflected in the petrography. Harper describes the voleanics of the Southern Coalfield as “a petrographical province of Permo-Carboniferous age possessing definite chemical characters and notably a high potash content.”5 In the series under discussion the alkalies are not high, and soda predominates. This marked difference can best be 3 Harper, L. F., Memoirs Geol. Sur. N.S.W., Geol. No. 7, p. 49. 4 Harper, L. F., op. cit., p. 307. 5 Harper, L. F., op. cit., p. 292. 6 Harper, L. F., op. cit., p. 278. 120 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. appreciated by comparing analyses representative of the rocks of the two areas. The analyses used for the Southern Coalfield area are those quoted by Harper on page 284 of his memoir. These are eight in number. It will be noticed that the highest silica percentage in Harper’s analyses is 59-64, while the majority of the flows in the Silverwood area are rhyolitic, and a typical analysis gives 73-93 percentage of silica. However, there is what in the light of modern petrogenesis must be considered an even more important chemical difference than the divergence in the average silica percentages, and that is the amount and proportion of the alkalies and lime present. In order to fairly illustrate this difference, the three analyses from the Silver- wood-Lueky Valley volcanics which bear the closest resemblance in silica content to the Southern Coalfield rocks are compared with the eight available analyses from the latter area. Extreme Values of Oxides of three Analyses from Silverwood — Lucky Valley. Extreme Values of Oxides of eight Analyses from Southern Coalfield. Si02 53-34-55-49 51 -07 -59.64 A1203 18-63-19-46 15-18 — 18-82 Fe203 1-72- 3-16 2-45- 7-30 FeO 4 27- 6-65 2-97— 5-40 MgO 2-75— 5-10 1-66- 4-09 CaO 5-81— 8-75 3-88- 7-72 Na20 2-30— 411 3-10- 3-97 K90 0-11- 0-62 2-75- 5-88 h~2o+ 1-03- 2-19 1 -07 — 2-89 Ti02 1-70— 1-82 0-42- 1-20 p2o5 014- 0-21 0-34- 0-88 The most striking difference in the two series is seen in the values for K20. The large amount of this oxide in the Southern Coalfield gives rise to the curious orthoclase- basalts and allied types in that area, which are in marked contrast with the volcanics with which we are concerned. (c) The Drake Area. — The Drake area is of interest in the present connection for several reasons. It is the closest area which contains late Palaeozoic volcanics comparable with those of the Silverwood-Lucky Valley district, being only some 50 miles distant. Lithologically the rocks of the VOLCANIC ACTIVITY IN SOUTHERN QUEENSLAND. 121 two areas have several points in common, and although the volcanics of the two areas appear to be somewhat different in age there is a parallelism in the sequence of their lavas. The volcanics of the Drake area have been described in considerable detail by Andrews.7 They “are referable to two distinct periods — an older, consisting mainly of rhyolites, and a younger, mainly of andesites.” These in turn are succeeded by the Drake fossiliferous slates and tuffs, but “the whole, felsites and slates, were shown to belong to one great period of sedimentation.”8 Andrews further states that the fossils in the covering slates and tuffs “are of marine types and belong to the Lower Marine division of the Permo-Carboniferous period.”9 An early Permo-Carboniferous age thus seems to be indicated for the volcanics of this area. The authors have examined specimens of the earlier Rhyolite Phase. These, while generally similar to the rhyolites of Silverwood-Lucky Valley, differed from them in the possession of veins of colloidal silica and in being somewhat mineralised (pyrites being fairly common). As neither rock analyses nor petrographica.1 descriptions have been published, more detailed comparisons are at present impossible. (d) The Berseker Raviges, near Rockhampton. — Through the courtesy of Mr. C. C. Morton, of the Queens- land Geological Survey, the authors have had an opportunity of studying hand specimens regarded by Mr. Morton as typical of the great series of volcanics forming an important part of the Berseker Ranges. The rocks are closely comparable with those typical of the Silverwood- Lucky Valley volcanics. Mr. Morton believes them to be partly intrusive and partly volcanic, and states that they rest upon and intrude the Rockhampton series of Car- boniferous age. Further evidence as to age is not forth- coming, but this close association of the volcanies with the Rockhampton series is suggestive of a Late Palaeozoic age. (e) The Bowen River Coalfield. — On the Bowen River Coalfield volcanic agglomerates associated with con- 7 Andrews, E. C., Min. Res. N.S.W., No. 12, p. 11. 8 Andrews, E. C., op. cit., p. 9. 5 Andrews, E. C., op. cit., p. 14. 122 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. glomerates are developed, but these are very low down in the series. Dunstan regards them as belonging to the Lower Marine division of the Permo-Carboniferous, while Jensen believes them to be even older. In addition to these there is evidence of volcanic activity in this area in the shape of basalts interbeddecl with the Upper Marine strata. The age of these corresponds with that of the volcanics under discussion, but very little petrological evidence is available for purposes of comparison. (/) The Mackenzie Fiver Area. — In the Mackenzie River area, also, basalts interbedded with Upper Marine strata are known. Andesites, the horizon of which has not been definitely established, are found both in this and in tho Dawson River areas. (V.) EARTH MOVEMENTS AND IGNEOUS ACTIVITY. The history of the Silverwood area may be briefly summarised as follows: — In Middle Devonian times exten- sive submarine outbursts resulted in the accumulation of a great thickness of andesitic tuffaceous material. This was immediately followed by the deposition of material which ultimately became a series of banded radiolarian cherts and mudstones. Pronounced orogenic movements in Late Devonian times (the Kanimbla Epoch of Sussmilch) compressed this “Silverwood Series” into isoclinal folds and elevated the Silverwood district into a land area, which it remained throughout the Carboniferous Period. During the latter part of this period very considerable igneous activity was in progress further to the south, in the Newr England district, but if the Silverwood area was also the scene of such activity there is no evidence of it. At the close of the Carboniferous period, and doubtless as the result of wRat in the New England area Sussmilch and David have described as the “Hunterian disturbance,”10 the Silverwood area was depressed below sea-level, and fossiliferous strata equivalent to the Lower Marine of the Hunter River district wrere deposited. These seem normal sediments, for the most part laid down under shallow water, and show no signs of violent contemporaneous volcanic activity such as that wdiich accompanied the accumulation of sediments of this age in the- Drake area some 50 miles to 10 Proc. Roy. Soc. N.S.W., 1919, p. 282. VOLCANIC ACTIVITY IN SOUTHERN QUEENSLAND. 123 the south. A slight change in sea-level resulted in the -deposition of freshwater sediments (equivalent to. the Greta Measures). Following this the area seems to have been subjected to a series of oscillatory movements, with the result that while the major portion of the sediments deposited are marine in nature (though of very shallow water types), there are intercalated with them at least two bands of freshwater deposits. The net result of these -oscillations was a gradual change from marine conditions to terrestrial conditions, and it was during this change that the volcanic activity which is the subject of this paper was manifested ; for while a considerable proportion of the lavas .and tuffs are found associated with fossiliferous marine ^strata, the major portion of them shows evidence of having been poured out on a land surface. No further marine deposits have been laid down in the area from this time to the present day. At some time subsequent to the deposition of the Upper Marine sediments and the associated volcanics, and ante- cedent. to the deposition of the Ipswich Series (of Upper Triassic age), there were intruded the huge well-known granitic masses of New England and Southern Queensland. The representative of this group in the Silverwood area is the coarse acid Stanthorpe granite. This might, perhaps, be expected to show some evidence of relationship with the volcanics, but a comparison of chemical analyses shows important points of difference, for whereas the alkalies of the Stanthorpe granite are approximately equal, there is in the volcanic rocks a normal soda value which predominates over an abnormally low value for potash. If we go a little further afield, however, we meet, at Greymare and elsewhere, with granitic rocks of a somewhat earlier phase than the acid Stanthorpe granite, which must therefore approximate even more closely in age the Permo- Carboniferous volcanics. One of us has endeavoured11 to show that the late Palaeozoic granitic rocks of Southern Queensland can be divided into an earlier ‘ ‘ Grey Phase, ’ ’ characterised chemically by a marked predominance of soda •over potash, and a later “Pink Phase,” where the alkalies are approximately equal or have potash in excess of soda. 11 Bryan, W. H., Proc. Roy. Soc. Qld., 1922, p. 148. 124 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. The earlier “Grey Phase” thus seems to be most nearly related to the volcanics under discussion, both in point of time and in chemical composition. The facts appear, then, to warrant the tentative assumption that the Permo- Carboniferous volcanics are genetically related to the granitic rocks of the “Grey Phase” as exemplified by the “ Hornblendic, Dioritic, and other basic granites” of New England12 and the “Maryland” and “Greymare” granites of the Stanthorpe area. ACKNOWLEDGMENT. The authors are indebted to Mr. R. A. Wearne, B.A., Principal of the Central Technical College, Brisbane, for providing facilities for the final drawing of the geological sketch section. 12 Andrews, E. C., Records Geol. Sur. N. S.W., vol. viii. , part 3, p. 212, M z tU X O’ iU E ^0 T3 O O O' & b X A— c s ,o c £ c D O 0 (Q £ o c o X- o vO UJ U I 3f *6 \- y O c o _o a m ill c L. O 5; t_ y a CL □ Q_ CL 5 § O tai 'N 0 tv. o o <0 p, o o 8-n a ^ o o > o (125) 126 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. PLATE IV. Figure 1. — Rhyolite from Eight Mile Range, showing flow structures Xi- Figures 2 and 3.— Rhyolites from Rhyolite Range, showing spherulitic structures xf. Proc. Hoy. Soc. Q’land, Yol. XXXV. Plate IV. Fig. 1. Fig. 2. Fig. 3. Face page 126. QUEENSLAND 1N0CERAMI WHITEHOUSE. 127 The Queensland Inocerami Collected by M. Lumholz in 1881. By F. W. Whitehouse, B.Sc., Foundation Travelling Scholar, Department of Geology, University of Queensland. (Plates V.-VII.) {Bead before the Royal Society of Queensland , 29th October, 1923.) Introduction. — The fossils about to be described constitute the collection of Inocerami made by Carl Lumholz from Minnie Downs, near Tambo, Queensland, in 1 881, and subsequently deposited in the geological museum of the University of Kristiania. Several of the specimens were later transferred to the University of Lund. I would take this opportunity of expressing my gratitude to Professors Kiaer, of Kristiania, and Gronwall,, of Lund, for their kindness in placing the whole of the collection at my disposal. The examination of this material was undertaken in connection with a general revision of the Queensland Inocerami, the results of which are to be published some time later. Owing to the fact that the present forms represent a definite collection lodged outside Australia, it was deemed advisable to publish the following account separately. A redescription of the type of /. maximus Lumholz was found necessary since the account given by Lundgren1 contains a very serious error. The large specimen which constitutes the type of I. maximus was found weathered out from the matrix. The other specimens, however, were encased in a block of stand- stone which was subsequently sawn in two. The Kristiania portion, when it reached me, measured 14 ins. x 7 ins. x 6 in., and was crowded with large specimens of Inocerami with both valves in apposition in nearly every case. By permission a large number of these forms were removed from the matrix for description in this paper. 1 B. Lundgren: “On an Inoceramus from Queensland.” Bihang, till. K. Sven. Vet.-akad. Handl., B. II. No. 5, 1885, pp. 3-6, pi. I. 128 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Description of the Species— The specimens examined form a related series characterised by (1) very slight inequality of the valves, (2) large size, (3) prominent pallial line and small sub-circular muscle scars, and (4) amazingly thin, relatively smooth shells. The type specimen of I. maximus measures 30 cms. along its greatest axis, yet its shell was only 1-5 mm. thick. A specimen of /. scutulatus attains a height of 10 cms. with a shell of 0-5 mm. thickness. Muscle scars are a feature extremely rarely seen in this genus, so that the presence of a prominent scar and pallial line on every specimen in this collection is of great interest. INOCERAMUS MAXIMUS Lumholz. (PL VII., figs. 1, 2 a, and 2b.) 1885. Inoceramus sp. B. Lundgren, Bihang. till. K. Sven. Vet.-aTcad. Handl., B. II., pp. 3-6, pi. I. 1889. I. maximus. C. Lumholz, “ Among Cannibals ” (London), p. 367, with text-fig. 8p. Chars. — Shell attaining a large size, the outline being that of an oblique oval with a slightly curved axis. Somewhat inequivalve ; very inequilateral, the unibones being terminal and the antero-dorsal area very sharply truncated. Postero-ventral angle regularly rounded. Height approximately equal to length ; hinge line about one- third of the total length. Umbones acute, the left slightly more prominent than the right. Axis of growth gradually curving concave to the hinge line. Pallial area very marked, roughly parallel to the outline and occupying about one-third of the internal surface. Shell substance extremely thin. Ornamentation consists of concentric ribs, which are not very marked on the superior part of the shell and almost disappear on the ventral portion. Ob s'. — The type specimen is a huge shell with the valves in apposition. Its dimensions are : Length 24 cms. ; height 23 cms.; width (two valves together) 8-5 cms.; hinge-line about 7 cms. ; thickness of shell substance 1*5 mm. This specimen was originally described and figured by Lundgren in 1885. 2 Unfortunately in doing so he mistook 2 Lundgren, Bihang, till. K. Sven. Vet-alcad. Handl., B. II., pp. 3-6, pi. I. QUEENSLAND INOCEBAMI WH1TEHOUSE. 129 the antero-dorsal margin for the hinge line, thereby con- fusing the terms left and right, and posterior and anterior. The description is thus very misleading; and the figure accompanying it, being orientated according to this error, gives a mistaken idea of the shell. It has therefore been necessary to redescribe it here. Lundgren’s figure was of the left valve, and should be turned (in an anti-clockwise direction) through an angle of 80° for correct orientation. The right valve has been figured in this paper. No specific name was given in the original description, but Lumholz3 reprinted Lundgren ’s plate in his book, and labelled it Inoceramus maximus , which name was adopted by Etheridge Jr.,4 and has been retained here. Etheridge endeavoured to identify specimens with this species relying on Lundgren’s work, but owing to the original mistakes referred to above such identifications cannot be maintained. INOCERAMUS SCUTULATUS sp. nov. (PI. V., fig. 1 ; pi. VI., fig. 1.) Sp. Chars. — Shell inequilateral with a lozenge-shaped outline modified somewhat by the rounding of the angles. Height slightly greater than length. Hinge line very short, about a quarter length of the shell. Shell expanded almost equally on either side of the axis of growth. Umbones acute, terminal, and prominent. Pallial area sac-shaped, lunate, occupying about a quarter of the internal surface. Shell substance extremely thin, ornamented by concentric ribs which are not very prominent. Axis of growth at right angles to the hinge line, and with a slight tendency to curve towards the posterior. Ligament pits small. Ohs. — This species is distinguished by its lozenge- shaped outline and the sub-equal anterior and posterior expansions. INOCERAMUS SCUTULATUS var. (PI. V., fig. 2; pi. VI., figs. 2 and 3.) There are a number of forms, three of which are figured here, which, while differing among themselves, lie on a line of variation linking the two species I. scutulatus and I. 3 Lumholz: “ Among Cannibals” (London), p. 367. 4 R. Etheridge, Jr.: Geol. Surv. Q’land. Bull., 13, 1901, pp. 24-25. e.s. — k. 130 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. maximus. The variation (considered, for the sake of description, as if proceeding from scutulatus towards maximus) begins in a rounding of the anterior and posterior median angles, thereby destroying the scutulate outline. At the same time the anterior portion of the shell decreases and the posterior tends to increase • while the growth oblique to the hinge line becomes more marked. I have refrained from attaching a definite varietal name to cover these forms; for, with the available information, and especially considering the lack of knowledge of the stratigraphical distribution, I cannot make up my mind how to place them. Two methods suggest themselves as being the most probable — (1) That these forms linking the two extremes of I. scutulatus and I. maximus form a variety of the same kind as I. concentricus var. subsulcatus 5 (which links the two forms I. concentricus and I. sulcatus ) ; or ('2) That the definition of I. scutulatus should be extended to cover forms like pi. VI., fig. 2, and 7. maximus should similarly be enlarged to include forms such as pi. V., fig. 2, and pi. VI.,. fig. 3. Thus until further information is available I prefer to leave the intermediate forms in an unnamed variety inter- mediate between the two species as defined above. XNOCERAMUS PROCERUS sp. nov. (PL VL, fig. 4.) Sp. Chars. — Shell oval, erect, inequilateral, slightly in equi valve. Height much greater than length (ratio 3:2). Hinge line about half the length of shell. Umbones terminal, acute, very prominent. Anterior portion sharply truncate, posterior flattened and sub-alate. Ornamentation (in adult specimens) consists of widely separated concentric ribs only on the earlier parts of the shell, the later parts, being smooth. Test very thin. Ligament pits small. Pallial area occupying about one-third of the internal surface. 5 See H. Woods: “The evolution of Inoceramus in the Cretaceous period,” Q.J.G.S. LXVIII., 1912, p. 4. QUEENSLAND INOCERAMI WHITEHOUSE. 131 Obs. — This species is distinguished by its erect habit. The hinge line is longer than in either of the two species described above. The area within the pallial line is about the same as in 7. maximus, but the direction of the axis of growth and only slight axial curvature ally it rather to 7. scutulatus. Probably both this species and 7. maximus- represent off-shoots from 1. scutulatus. Comparison with other Species. — 7. scutulatus may be closely compared with the short form of 7. crippsi var. reoxhensis Eth.6 from the Cenomanian of England. Both species have the subequal anterior and posterior expansions of the valves, and there is a certain scutulate outline, but much less pronounced, in the latter variety. It differs from, the English form in having the umbones more prominent, the hinge line shorter, and the concentric ornamentation much less pronounced. 7. maximus shows a pronounced resemblance to the widespread 7. labiatus7 (Schloth.) (Cenomanian and L. Turonian), which is also a large form. The resemblance is most apparent in the direction of growth oblique to the hinge line (a characteristic feature of 7. labiatus). It differs from this form particularly in the more-acute umbones and the less-prominent ornamentation. Eichwald8 figures a specimen labelled 7. mytiloides from the Cretaceous of Russia, which very closely resembles the original of pi. V., fig. 2, of this paper. (7. mytiloides, it is to be noted, is regarded by Woods as a synonym of I. labiatus.) Again, the Queensland form differs in the more- pronounced umbones and less-prominent ornamentation. Woods has shown9 for the English forms that 7. labiatus has most probably evolved direct from I. crippsi var. reachensis, and it is thus interesting to note that I. scutulatus and 7. maximus, which are themselves connected by intermediate forms, are individually comparable with 6 See IT. Woods: “Monogr. Cret. Lamellibr.” (Palseont. Soc.), vol. II., 1911, p. 278, pi. XL VIII., fig. 5; pi. XLIX., fig. 1. 7 See Woods, op. cit., pp. 281-284, pi. L. 8 Eichwald, “Lethsea Rossica’ ’ (per. moyenne), pi. XXI., fig. 6. “Woods, ‘ ‘ The evolution of Inoceramus in the Cretaceous period, Q.J.G.S. LX VIII., 1912, p. 13. 132 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND these two English species. I cannot regard the Queensland forms as identical with the European species, for all specimens show the consistent distinctive features noted above. It remains to be seen whether these species will show the same stratigraphieal distribution as the European forms. Further comparisons are held over until, in a later paper, the Queensland Inocerami are treated as a whole. But it may be noted that in these particular species I find no very close relationships with Indian or New Zealand forms. These collections return, of course, to the Universities of Kristiania and Lund ; but plaster casts of the types will be brought to Queensland. Proc. Roy. Soc, Q*land, Vol. XXXV. Plate "V. Fig. 1. — Inoceramus scutulatus Whitehouse. Right valve of type specimen, x t (Kristiania collection). Fig. 2. — Inoceramus scutulatus var. Right valve of a specimen closely approaching I. maxim, us, X f (Kristiania collection). ''ace page 132 Proc. Koy. Soc. Q’land, Yol. XXXV. Plate VI. Fig. 1. — Inoceramus scutulatus Whitehouse. Anterior view of PI. V., fig. 1, X f. Figs. 2 and 3. — Inoceramus scutulatus var. Two small right valves, X f. Fig. 2 approaches the normal type of I. scutulatus. Fig. 4. — Inoceramus procerus, Whitehouse. Left valve of type specimen, approx. X h Face page 132 • Proc. Roy. Soc. Q ’land, Yol. XXXV. Plate YII. Inoceramus maximns Lumholz. Fig. 1 — Right valve of type specimen, X § (Kristiania collection). Fig. 2 — Smaller specimen, X f (Lund collection); a, left valve; ft, antero-ventral view. Face vage 132. VOLATILE OIL OF DAPHNANDRA AROMATICA. 133 The Composition of the Volatile Oil of the Leaf of Daphnandra aromatica Bailey. By T. G. H. Jones, B.Sc., A.I.C., and Frank Smith, B.Sc., F.I.C. {Bead before the Boyal Society of Queensland, 26th November, 1923.) Subsequent to the publication of onr note (this volume, pp. 61-2) on the essential oils from the bark and leaves of Daphnandra aromatica, we received from North Queensland through the courtesy of Mr. E. H. F. Swain, Director of Forests, a further supply of leaves permitting of chemical examination in detail of their volatile oil. Contrary to expectation from the cursory examination of the small amount of leaf oil (40 ccs.) previously avail- able, limonene was not found to be present in the lower boiling fractions of the oil. The principal constituents of the oil are da-phelland- rene and a sesquiterpene; the other bodies present being d-pinene, cineol and a sesquiterpene alcohol together with small quantities of an aldehyde (probably iso-valeric or caproic) and a phenolic body. In view of the present inexact condition of knowledge of the sesquiterpene group and of the scantiness of the data secured, it is not possible to decide whether the sesquiter- pene and the sesquiterpene alcohol of Daphnandra leaf oil are each distinct chemical entities or are identical or not. with any similar bodies previously described. Experimental. — The distillation in steam of 200 lbs. of air-dried leaves yielded 400 ccs. or -5 per cent, of a pale- yellow oil of agreeable odour possessing the following constants : — ] Specific gravity 15;5- -9084 (N)D20 1-4892 {a) D + 23-6. Ester value 10, Acetyl value 40. 134 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Five ccs. of the oil were extracted in turn with sodium bisulphite and with sodium hydroxide solutions. Absorp- tion to the extent of about 1 per cent, of an aldehydic body and -5 per cent, of a phenolic body giving* a greenish colour with ferric chloride were recorded. On fractionation of 370 ccs. of the oil at 2 mms, pressure, using an oil pump with a trap cooled in liquid ammonia interposed between the pump and receiver, there were collected — (1) In ammonia trap (2) Below 50° C .. (3) At 50- 60° C (4) At 60- 70° C (5) At 70- 80° C (6) At 80-100° C (7) At 100-115° C 1 8 ) At 115-128° C 10 ccs. 91 ccs. 46 ccs. 15 ccs. 20 ccs. 108 ccs. 25 ccs. 40 ccs.. leaving a small viscous residue in the distilling flask. The presence of cineol was indicated in the lower fractions by its odour, and fractions (2), (3), (4), (5), and (6) were accordingly extracted several times with 50 per cent, resorcin solution. The unabsorbed oil in these fractions was then refrac- tionated at 21 mms. pressure, and the following fractions ultimately collected (9) Boiling below 65° C. 15 ccs. Sp. Gr. -861 (N)D20 1466 (a)D + 50 (10) Boiling at 65°-68° C. 60 ccs. Sp. Gr. -8538 (N)D20 14745 (a)D + 59 d-Pinene. — The physical constants as well as the odour of fraction (9) suggested the presence of pinene, although The presence of phellandrene was also indicated by the nitrosite reaction. It did not, however, yield the nitrosyl chloride characteristic of pinene (owing, no doubt, to its high degree of optical activity), but the presence of pinene was confirmed by oxidation with potassium permanganate solution. Five ccs. of the oil were shaken with 10 grammes of potassium permanganate and 150 ccs. of water and 100 VOLATILE OIL OF DAPHNANDRa AROMATICA. 135 grammes of ice until the permanganate was completely reduced. The liquid was then filtered from the manganese dioxide and evaporated to small bulk. Small amounts of neutral oxidation products were extracted with ether and the liquid then acidified with dilute sulphuric acid and again extracted with ether. On evaporation of the ether the remaining liquid (about 4 ccs.) was identified as pinonic acid by means of its -semi-carbazone melting at 204° C, and pinene was therefore present in the fraction. The fraction (1) collected in the ammonia trap also consisted largely of pinene, but the presence of an aldehyde was indicated by Sehiff’s Reaction. Attempts were made to extract it by means of semi-carbazide hydrochloride, but the small amount of material obtained prevented purifica- tion of the resulting semi-carbazone. The irritating and unpleasant odour, however, would suggest identity - with isovaleric or caproic aldehyde. ck-Phellandrene. — Fraction (10) consisted largely of da-phellandrene. It was identified as such by means of its nitrosite prepared in accordance with the method described by Smith, Hurst, and Read (J.C.S. Transactions, 1923, 1657). The resulting nitrosite melted at 120-121° C. Cineol. — The oil extracted by the resorcin solution (described above) was recovered. by distillation in steam, and found to consist almost exclusively of cineol. It was identified as such by its physical constants Sp. Gr. 9330, (N)d2o 1-458, as well as by its characteristic odour. Sesquiterpene. — It was evident from analysis that fraction (6) consisted largely of a sesquiterpene. It was refractionated and finally redistilled repeatedly (at the reduced pressure) over metallic sodium. B.P. 110-111°C (4 mms.) 146-I48°C (24 mms.). (Found C J 87-7%. H = 1T4%. C15H24 requires C = 88*2. H=1T8). Sp. Gr. -9106 (N)D20 T5012 (a)D + 2*7. Molecular refraction 66 (calculated 62*6 for C15 H24). 136 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. None of the solid derivatives used to characterise many known sesquiterpenes could be prepared. A liquid hydro- chloride was obtained by passing dry HC1 gas into its ethereal solution (cooled with ice), and the hitrosyl chloride and nitrosite were also apparently liquid. The molecular refraction indicated probable di-cyclic character, and that it contained two double bonds. Treated with a few drops of glacial acetic acid and bromine vapour a violet colour was developed, rapidly, turning deep-blue. Sequiterpene Alcohol. — Fraction (8) w^as a viscid liquid of a light-green colour. Sp. Gr. -9666 (N) D20 about 1-50. Boiling point 170-180° C (24 mms.). Its alcoholic character was demonstrated by its action on metallic sodium and reaction with acetic anhydride. Summary. — The air-dried leaves of Daplmandra aroma- tica yielded on distillation with steam -5 per cent, of a light- yellow oil possessing an agreeable aromatic odour, the com- position of which is indicated as approximately d-|)inene 5-10 per cent., cla-phellandrene 20-25 per cent., cineol 10-15 per cent., sesquiterpene 30-40 per cent, sesquiterpene alcohol 5-10 per cent., with minor aldehydic phenolic and ester constituents. THE, ROYAL SOCIETY OF QUEENSLAND ABSTRACT OF PROCEEDINGS Report of Council for 1922. To the Members of the lioyal Society of Queensland. Your Council has pleasure in submitting its Report for the year 1922. During the year twelve papers were published. With a few exceptions, these papers were read before the Society. In addition, the following lectures, which were well attended and to which the public was invited, were delivered : 1 ‘ The Geology of Northern Australia,” by Dr. H. I. Jensen; “Kosciusko, the Roof of Australia,” Professor H. C. Richards; “The Eucalypts of the Brisbane District,” C. T. White; and “Corals,” Professor F. Wood-Jones. The last-mentioned lecture was held in conjunc- tion with the Royal Geographical Society of Queensland. The Society is indebted to the University of Queensland for- providing accommodation for meetings and for housing the library. We wish to acknowledge the practical assistance afforded the Society by the Government of Queensland, who, as in previous years, voted £50 towards the Society’s activities. Appreciative acknowledgment is also accorded to the Govern- ment of Papua for a subsidy of £25 towards the printing of Mr. C. T. White’s paper on the Flora of Papua and to the Walter and Eliza Hall Fund for subsidies towards the publi- cation of the following papers: “Notes on the Biology of Some of the More Common Queensland Muscoid Flies,” by T. H. Johnston and 0. W. Tiegs, “New and Known Sarcophagid Flies” by T. IT. Johnston and 0. W. Tiegs, and “A Synonymic List of Some Described Calliphorine Flies” by T. IT. Johnston and G. H. Hardy. VI ABSTRACT OF PROCEEDINGS. The membership roll consists of 84 ordinary members, 9 life members, 10 corresponding members, and 3 associate members. During the year 8 new members were elected, and one member resigned.. With deep regret we record the deaths since last annual meeting of two of our members, namely, Dr. A. Sutton, C.B., C.M.G., and Mr. J. Johnston. Eight meetings of the Council were held during the year. The attendance was as follows: IT. J. Priestley (President) 7, E. W. Bick 8, W. H. Bryan 6, W. D. Francis 7, E. H. Gurney 6, T. IT. Johnston 3, H. A. Longman 5, E. 0. Marks 8, H. C. Richards 4, E. H. Swain 4, C. T. White 7. Attention is drawn to the accompanying Statement of Receipts and Expenditure, which shows that the Society’s funds are almost exclusively devoted to printing. H. J. PRIESTLEY, President. W. D. FRANCIS, lion . Secretary. AUSTRALASIAN ASSOCIATION FOR THE ADVANCEMENT OF SCIENCE. The Society was represented by Professor IT. C. Richards, D.Sc., at the meeting of the Australasian Association for the Advancement of Science held at Wellington from 10th to 18th January, 1923. In the report which Professor Richards submitted to the Council, it is stated that it was decided to have the next meeting of the Association at Adelaide in August, 1924. CORRIGENDUM. On page 2, Proceedings of this Society for 1922 (Vol. XXXIV.), in paragraph referring to the death of Dr. J. Shirley, 12th line from bottom of page, for “5th March” read “4th April.” ABSTRACT OF PROCEEDINGS. Vll r^CiOOOOOOOOCO soioooeacoooqi— ioco rH rH r— I r-H CR ^ O O Cd O Cd r-l Od Q Z < J 0) z 111 1x1 D O Ll O > h LlJ 0 0 If) J < > 0 (T LU 1 h y S’ ffl m O rA Cd ffl H ffl ffl L- « O oo lO CO O o o O rH O O Od rH o © =i CC •'f S S r9 ^ o d ^ ft > S - c3 O ft! - CL os. > QC ^ O ,g ° ft O ■+■> © ^ ft OJO co ft ’£ ft? ft CC o © © ft ^ rrt © g £ © © So* o £ ^ H fto o P; © m O' i GQ Examined and found correct. H. J. PRIESTLEY, E. W. BICK, Hon, Auditor, 27th February, 1923. Treasurer. Vlll ABSTRACT OF PROCEEDINGS. Abstract of Proceedings, 26th March, 1923. The Annual Meeting of the Society was held in the Geology Lecture Theatre of the University at 8 p.m. on Monday, 26th March, 1923. His Excellency Sir Matthew Nathan, P.C., G.C.M.G.,, Patron of the Society, presided. The minutes of the previous Annual Meeting were confirmed. Drs. R. W. Cilento and S. Fancourt McDonald were unanimously elected as Ordinary Members. Archbishop Duhig, Professor E. J. Goddard, B.A.,„ D.Sc., Professor J. P. Lowson, M.A., M.D., G. P. Dixon, C.B.E., M.B., Ch.M., J. G. Hamlin, M.Sc., Mrs, Hamlin, M.A., Miss Barker, B.A., were proposed as Ordinary Members, and J. PI. Simmonds, B.Sc,, and F. G. Holdaway, B.Sc., were proposed as Associates. On the motion of Prof. Richards, seconded by Mr. Longman, the Annual Report of the Council and the Statement of Receipts and Expenditure were adopted. The following officers were elected for 1923 : — President: Dr. E. 0. Marks, M.D., B.A., B.E. Vice-Presidents: Prof. H. J. Priestley, M.A. {ex- officio) ; W. H. Bryan, M.Sc. Patron: His Excellency Sir Matthew Nathan, P.C.r G.C.M.G. Hon. Secretary: W. D. Francis. Hon. Treasurer: E. W. Bick. Hon. Editor: H. A. Longman, F.L.S. Hon. Librarian: W. H. Bryan, M.Sc. Hon. Auditor: Prof. H. J. Priestley, M.A. Members of Council: E. H. Gurney, Professor H. 0~ Richards, E. H. Swain, R. A. Wearne, B.i.„ and C. T. White, F.L.S. The retiring President, Prof. PI. J. Priestley, M.A., delivered his presidential address entitled, “On the* Development of Scientific Thought. ” A cordial vote of thanks to His Excellency the Governor was carried by acclamation. ABSTRACT OF PROCEEDINGS. IX Abstract of Proceedings, 30th April, 1923. The Ordinary Monthly Meeting of the Society was held in the Geology Lecture Theatre of the University at 8 p.m. on Monday, 30th April, 1923. The President, Dr. E. 0. Marks, M.D., B.A., B.E., in the chair. The minutes of the previous Monthly Meeting were read and confirmed. Drs. J. Lockhart Gibson, T. W. H. Mathewson, and E. N. Merrington, Rev. C. H. Massey, and Mrs. E. Lnmley Hill were nominated for ordinary membership. Archbishop Duhig, Professor E. J. Goddard, B.A., D.Sc., Professor J. P. Lowson, M.A., M.D., G, P. Dixon, C.B.E., M.B., Ch.M., J. G. Hamlin, M.Sc., Mrs. Hamlin, M.A., Miss Barker, B.A., were elected as Ordinary Members, and J. H. Simmonds, B.Sc., and F. G. Holdaway, B.Sc., were elected as Associates. In making an exhibit of specimens, drawings, and slides of Queensland fossil insects, Mr. B. Dunstan gave a general account of the insect-bearing beds of Denmark Hill, Ipswich District. These beds were discovered by Mr. J. H. Simmonds when looking for fossil plants. Many years afterwards they were carefully investigated by officers of the Queensland Geological Survey. Illustrations of very old types of the dragon-fly, mantis, locust, lacewing, cock- roach, bug, jassid, and beetle were screened. One slide showed a restoration of Ipsvicia Jonesi, a giant jassid, named after the Minister for Mines. This restoration was made possible by the recent discovery of an almost perfect specimen. The insect quarry at Ipswich has yielded about 120 new species of fossil insects, and no doubt others will be unearthed as the investigation proceeds. Mr. W. H. Bryan, M.Sc., communicated a paper by Dr. R. J. Tillyard, M.A., D.Sc., F.E.S., entitled "On a Tertiary Fossil Insect Wing from Queensland.” The paper describes a new insect discovered by Mr. Bryan in the Tertiary beds at Redbank Plains, near Goodna. One other fossil insect wing ( Euporismites balli) has been found in the same deposits, and described by Dr. Tillyard, while numerous fossil fish and dicotyledonous plants have been found on the same horizon. The new insect wing has been ABSTRACT OF PROCEEDINGS. X named Scolypopites bryani , n.g. et sp., the generic name indicating its close affinity to the recent Australian genus Scolypopa, the common passion-vine hopper of Eastern Australia, to which it is probably directly ancestral, and the specific name being a dedication to its discoverer. Prof. Richards, Mr. Tryon, Prof. Goddard, Dr. A. J. Turner, and Mr. Bryan took part in the discussion on the paper and Mr. Dunstan’s exhibits. Mr. G. H. Hardy read a paper by Prof. T. IT. Johnston and himself, entitled “Observations Regarding the Life Cycle of Certain Australian Blowffiies, ’ ’ and gave an account of some of the work he was engaged upon as a Walter and Eliza Hall Fellow. The work of the Fellowship consisted chiefly of experimental investigation of the biology of blow- flies. It had been discovered that a batch of these flies bred under uniform conditions varied greatly in the period of the life cycle, particularly in the prepupal stage, and it was pointed out that chalcid wasps with suitable habits in other parts of the world might be introduced into Australia to supplement those already preying upon the sheep maggot flies here. Abstract of Proceedings, 28th May, 1923. The Ordinary Monthly Meeting of the Society was held in the Geology Lecture Theatre of the University at 8 p.m. on Monday, 28th May, 1923. The President, Dr. E. 0. Marks, M.D., B.A., B.E., in the chair. The minutes of the previous Monthly Meeting were read and confirmed. J. Lockhart Gibson, M.D., T. H. R. Mathewson, M.B.,. Ch.B., E. N. Merrington, Ph.D., Mrs. E. Lumlev Hill, and Rev. C. H. Massey were elected as Ordinary Members. A paper by Mr. John C. Hamlin, M.Sc., entitled “New Cactus Bugs of the Genus Chelinidea (Ilemiptera),” was tabled and taken as read. Two new species and one new variety are described in the paper. These insects were collected by the author in North America for the purpose of endeavouring to control prickly-pear in Australia by biological agencies. ABSTRACT OF PROCEEDINGS. XI A paper by Dr. T. L. Bancroft entitled “Some Further Observations on the Dawson River Barramundi: Scleropages leichhardtii,” was communicated by the Hon. Secretary. The author concludes that the following facts have been ascertained Barramundi carries the spawn in its mouth ; October is the spawning season ; fish when meshed, eject the spawn; sexually mature fish are not meshed in a net of 3-inch mesh; fish secured in the bunt of a drag-net would retain the ova and young fish in their mouths. Mr. W. H. Bryan, M.Sc., read a paper entitled “An Unusual Tourmaline-albite Rock from Enoggera, Queens- land.” This curious rock, found by Mr. Bryan near the contact of the Enoggera granite with the Brisbane schist, is composed entirely of tourmaline and albite. It shows a pronounced schistose structure. The only other rock of the same constitution previously described is found in Cornwall, England. The President exhibited specimens of carbonaceous earth found at Samford where a bush fire had ignited it. The earth continued to burn for some week afterwards, and deposited a substance with the appearance of baked clay. The President, Messrs. F. Bennett, R. A. Wearne, and W. H. Bryan took part in the discussion on the papers. Abstract of Proceedings, 27th June, 1923. The Ordinary Monthly Meeting of the Royal Society of Queensland was held in the Geology Lecture Theatre of the University at 8 p.m. on Wednesday, 27th June, 1923. The President, Dr. E. 0. Marks, B.A., B.E., M.D., who was in the chair, announced that His Excellency, Sir Matthew Nathan, had expressed his regret at being unable to attend through indisposition. The minutes of the previous monthly meeting were read and confirmed. A paper by T. G. Jones, B.Sc., A.I.C., and F. B. Smith. B.Sc., F.I.G., entitled, “Notes on the Essential Oil of Xll ABSTRACT OF PROCEEDINGS. Daphnandra aromatica was tabled and taken as read. Chiefly the oil of the bark, which contains 95 per cent, of safrol, is dealt with. Mr. H. A, Longman, F.L.S., delivered a lecture entitled, ‘ 1 Australian Marsupials, ’ ’ illustrating his remarks with specimens and lantern slides. Although species of large macropods were at present so numerous as to be pests in many Western districts, a plea was made for fenced inland reserves— including one by the main railway — for their special protection. The lecturer stated that the fossil marsupial fauna was even more characteristically 'Australian than that of to-day, and the evolution of the group had largely taken place here. He thought that fossil evidence would one day be forthcoming for the northern origin of remote ancestors. A vote of thanks was accorded the lecturer on the motion of Dr. J. Y. Duhig, seconded by Mr. R. A. Wearne. Mr. H. Try on and Mr. Donald Gunn (a visitor) took part in the discussion. The Ordinary Monthly Meeting of the Society will be iheld in the Geology Lecture Theatre of the University at 8 p.m. on Tuesday, 24th July. Abstract of Proceedings, 24th July, 1923. The Ordinary Monthly Meeting of the Society was held in the Geology Lecture Theatre of the University at 8 p.m. on Tuesday, 24th July, 1923. The President, Dr. E. 0, Marks, B.A., B.E., M.D., in the chair. The minutes of the previous monthly meeting were read and confirmed. Mr. P. B. Guthrie, F.I.C., was nominated for Ordinary Membership. The Council’s appointment of Prof. E. J. Goddard, B.A., D. Sc., as Hon. Librarian was ratified on the motion of Mr. P. Sylow, seconded by Prof. Richards. ABSTRACT OF PROCEEDINGS. Xlll On the motion of Prof. Richards, seconded by Mr. Longman, the Society’s appreciation was accorded Mr. W. H. Bryan, M.Sc. (Vice-President), for his work as Hon. Librarian. Mr. H. A. Longman, F.L.S., exhibited two skins of pouch embryos of the grey kangaroo ( Macropus gigan - tens ), which had been taken from one mother by Mr. H. S. Smith of Morven, and sent to the Queensland Museum. One skin was quite naked, whilst the other was consider- ably larger and thinly haired. Although it was just possible that the kangaroo may have acted as foster mother to the larger pouch embryo, it seemed probable that this was a case of supplementary twins, such as that recorded in the Proceedings of the Zoological Society, 1912, page 234. Mr. C. T. White, F.L.S., read a paper, by himself and W. D. Francis, entitled “Contributions to the Queens- land Flora, Part II.” One new genus ( Placospermum , Nat. Order Proteaceae) and nine new species are described. The new species are Calopkyllum touriga, Aglaia ferru- ginea, Eugenia macrohila, Eugenia Petriei, Ardisia bifaria, Cryptocarya foveolata, Cryptocarya pleurosperma, Placos- permum coriaceum and Croton denser est it um. Sixteen species are recorded for the State for the first time. Mr. Longman discussed the paper. In communicating his paper entitled, “Notes on the Physiography of Eastern New Guinea and Surrounding Island Groups,” Rev. C. H. Massey delivered a lecture on the subject. He described the geological features of the Rabaul Harbour and the country within a 20-mile radius. The recent volcanic disturbances in that area were referred to, and the lecturer correlated the predominant mountain ranges of Eastern New Guinea and New Britain with the ocean basins of the South-Western Pacific. A vote of thanks was accorded the lecturer on the motion of Prof. Richards, seconded by Mr. W. H. Bryan, and supported by the President and Mr. L. C. Ball. Owing to the absence of a large number of the active members of the Society, the August Monthly Meeting was not held. r.s. — L. XIV ABSTRACT OF PROCEEDINGS. Abstract of Proceedings, 24th September, 1923. The Ordinary Monthly Meeting of the Society was held in the Geology Lecture Theatre of the University at 8 p.m. on Monday, 24th September, 1923. The President, Dr. E. O. Marks, B.A., B.E., M.D., in the chair. The minutes of the previous Monthly Meeting were read and confirmed. Dr. F. W. S. Cumbrae-Stewart, B.A., Prof. T. Parnell, M.A., and Mr. T. W. Heney were nominated for member- ship Mr. F. B. Guthrie, F.I.C., was unanimously elected as an ordinary member. Specimens of the Quetta “meteorite” were exhibited on behalf of Prof. Skertchly. The exhibit was commented upon by Prof. Bichards. Prof. H. C. Bichards, D.Sc., read a paper by himself and Mr. W. H. Bryan, M.Sc., entitled “Permo-Carboni- ferous Volcanic Activity in Southern Queensland.” A series of lavas and tuffs met with in the Silverwood-Lucky Valley area are dealt with in the paper. These rocks, on account of their resistance to weathering, are the dominating topographic forms in their immediate locality, and in some cases act as important divides. One outcrop measured 4,250 ft. iu minimum thickness, and the authors have reasons for assuming that the total thickness is much greater. Both the flows and the associated pyroclastic rocks are for the most part of an acid nature, and many beautiful examples of spheroidal and fluidal rhyolites were observed. More basic volcanics are represented in the upper part of the series, but in smaller amount. A vote of thanks was accorded the authors on the motion of Mr. L. C. Ball, B.E., seconded by Dr. H. I. Jensen, and supported by Mr. Gurney and the President. Abstract of Proceedings, 29th October, 1923. The Ordinary Monthly Meeting of the Society was held in the Geology Lecture Theatre of the University at 8 p.m. on Monday, 29th October, 1923. ABSTRACT OF PROCEEDINGS. XV The President, Dr. E. 0. Marks, B.A., B.E., M.D., in the chair. The minutes of the previous monthly meeting were read and confirmed. Dr. Cumbrae-Stewart, Mr. T. W. Heney, and Professor T. Parnell, M.A., were unanimously elected as ordinary members. Mr. W. B. Alexander, M.A., exhibited pellets produced by the Boobook Owl, “Laughing Jackass,” Wedge-tailed Eagle, and Australian Crow collected by him in the Rock- hampton district. He stated that though it had long been known that owls ejected from the mouth pellets of bones and fur, it had only recently been discovered that some Australian Kingfishers had a similar habit, and it had not hitherto been known that eagles and crows produced pellets. Those of the crow exhibited were of special interest because they contained numerous seeds of prickly- pear, thus indicating conclusively that these birds spread that plant. Professor E. J. Goddard, B.A., D.Sc., exhibited speci- mens of Peripatus capensis, which, he explained, was the classic subject worked upon by Professor Sedgwick. It is the finest species of the genus and is distributed in the West Indies, Brazil, South Africa, India, East Indies, Australia, and New Zealand. Like so many other forms of the Southern hemisphere it is a primitive type, and its distribution has been much abused by zoogeographers. The President exhibited specimens of hollow cal- careous nodules containing drusy cavities, which were found in sandstone near Ipswich, and a cylindrical stone implement of unusual shape, which had been ploughed up near Taroom, and sent in by Mr. A. H. Blackman. This would be placed in the Queensland Museum. Mr. W. H. Bryan, M.Sc., communicated a paper entitled, “The Queensland Inocerami, collected by Lum- holz in 1.881,” by P. W. Whitehouse, B.Sc., Foundation Travelling Scholar, Department of Geology, University of Queensland. In the paper the author describes the results of his re-examination of the Inocerami which were collected by Lumholz from Minnie Downs, near Tambo, in 1881, and subsequently deposited in the University of Kristiania. XVI ABSTRACT OF PROCEEDINGS. The examination of this material was undertaken in connection with a general revision of the Queensland Inocerami. The author redescribes the type of I. maximus Lumholz, describes two new species, I. scutulatus and I. procerus, and compares these with European and other species. Professor Richards commented upon the paper. Dr. A. Jefferis Turner, M.D., opened a discussion on the origin of the Australian Fauna, Professors Goddard and Richards, and Messrs. Longman, Bryan, Alexander, Bennett, and Hardy took part in the discussion. Two public lectures by Professor J. Cossar Ewart, F.R.S., of the Edinburgh University, were held under the auspices of the Society. One, entitled “The History of Feathers and the Breeding of King Penguins, ? 7 was held on 17th October. His Excellency Sir Matthew Nathan, P.C., G.C.M.G., presided. A vote of thanks was accorded the lecturer on the motion of Dr. J. Lockhart Gibson, M.D., seconded by Mr. W. B. Alexander, M.A. The other, entitled “The Scientific Breeding of Sheep,” was held on 18th October. The Premier of Queensland, the Hon. E. G. Theodore, presided. A vote of thanks was accorded the lecturer on the motion of Mr. A. H. Whit- tingham, seconded by Professor Goddard, B.A., D.Sc. Abstract of Proceedings, 24th November, 1923. The Ordinary Monthly Meeting of the Society was held in the Geology Lecture Theatre of the University at 8 p.m. on Monday, 24th November, 1923. The President, Dr. E. 0. Marks, B.A., B.E., M.D., in the chair. The minutes of the previous Monthly Meeting were read and confirmed. T. G. H. Jones, B.Sc., A.I.C., and P. C. Tibbits were nominated for membership. The President announced that as the result of a deputation to the Premier, the Hon. E. G. Theodore, the Government had consented to subsidise the Society to the ABSTRACT OF PROCEEDINGS. XVil extent of £ for £ of the amount expended in printing up to £150 per annum. The deputation consisted of Professors. Richards and Goddard and himself. Mr. H. A. Longman, F.L.S., exhibited a sub-fossil aboriginal mandible, dredged with sand and gravel from the Brisbane River near Indooroopilly Bridge and presented to the Queensland Museum by Dr. E. S. Jackson, which was. within the range of variation of present-day specimens. He also exhibited a young P ter opus poliocephalus (head and body 155 mm. in length), which was attached to its. mother when shot during flight. Mr. B. Dunstan, Chief Government Geologist, delivered a lecture entitled “Some New Ideas on the Artesian System.” His remarks were illustrated by a series of coloured lantern slides. Messrs. Tibbits, Prof. Richards, J. B. Henderson, Prof. Goddard, and the Presi- dent took part in the discussion on the subject. A paper by T. G. Jones, B.Sc., and P. B. Smith, B.Sc.,. entitled “The Composition of the Volatile Oil of the Leaf of Daphna?idra aromatica. , ” was tabled and taken as read. XVlll PUBLICATIONS RECEIVED. Publications have been received from the following Institutions Societies, etc., and are hereby gratefully acknowledged. AFRICA. Durban Museum, Durban. Brazil — AMERICA. Institute* Oswaldo Cruz, Rio Janeiro. Canada — - Department of Mines, Ottawa. Royal Canadian Institute, Toronto. Royal Astronomical Society of Canada, Toronto. Royal Society of Canada, Ottawa. United States — American Academy of Arts and Sciences, Boston. American Geographical Society, New York. Academy of Natural Science, Philadelphia. American Philosophical Society, Philadelphia. Californian Academy of Science, San Francisco. Indiana Academy of Science. Rochester Academy of Science. National Academy of Science and Smithsonian Institute, Washington Library of Congress. American Museum of Natural History, New York City. Newr York Zoological Society, New York. Department of Agriculture. Arnold Arboretum. Department of Commerce, New York. Missouri Botanic Gardens, St. Louis, Missouri. University of California, Berkley. University of Colorado. Cornell University. John Hopkins University. University of Illinois, Urbana. Ohio State University. University of Kansas, Lawrence. University of Minnesota, Minneapolis. Oberlin College. National Research Council. Buffalo Society of Natural Science. Portland Society of Natural History. Lloyd Library. The University of Michigan, Michigan. Bernice Pauahi Bishop Museum, Honolulu, Hawaiian Islands. PUBLICATIONS RECEIVED. XIX Mexico — Institute) Geologico d'e Mexico, Mexico. Sociedad Cientifica, Mexico. Colombo Museum, Colombo, Ceylon. Agricultural Institute, Pusa, Bengal. Geological Survey of India. Superintendent, Government Printing. Survey of India, Delira Dun. Java — Department van Landbrouw, Batavia. Konincklyke Naturkundige, Batavia. Philippine Islands — Bureau of Science, Manila. AUSTRALIA AND NEW ZEALAND. -Queensland — Field Naturalists ’ Club, Brisbane. Geological Survey of Queensland, Brisbane. Queensland Museum, Brisbane. Government Statistician, Brisbane. New South Wales — Australasian Association for the Advancement of Science, Sydney. Department of Agriculture, New South Wales. Botanic Gardens, Sydney. Geological Survey of New South Wales, Sydney. Public Library, Sydney. Linnean Society of New South Wales, Sydney. Australian Museum, Sydney. Royal Society of New South Wales, Sydney. Naturalists’ Society of New South Wales, Sydney. University of Sydney. Victoria — Bureau of Census and Statistics, Melbourne. Royal Society of Victoria, Melbourne. Field Naturalists’ Club, Melbourne. Department of Agriculture, Melbourne. Department of Mines, Melbourne. Australasian Institute of Mining and Metallurgy, Melbourne. National Museum, Melbourne. Institute of Science and Industry, Melbourne. Scientific Australian. XX PUBLICATIONS RECEIVED. Tasmania — Royal Society of Tasmania, Hobart. Field Naturalists’ Club, Hobart, Tasmania. Geological Survey of Tasmania. South Australia — Royal Society of South Australia, Adelaide. Royal Geographical Society of South Australia, Adelaide. National Museum of South Australia, Adelaide. Geological Survey of South Australia, Adelaide. West Australia — Royal Society of West Australia, Perth. Geological Survey of West Australia, Perth. New Zealand — Auckland Institute, Auckland. New Zealand Board of Science and Art. Dominion Laboratory, Wellington. Geological Survey of New Zealand, New Zealand Institute, Wellington. EUROPE. Austria — Annals Natural History Museum, Vienna. Belgium — Acad’emie Royale, Brussels. Societe Royale de Botanique de Belgique. Societe Royale de Zoologique de Belgique. Czecho-Slovakia — Spolecnosti Entomologicke, Prague. Great Britain — Cambridge Philosophical Society. Conchological Society of Great Britain and Ireland. Imperial Bureau of Entomology, London. Literary and Philosophic Society, Manchester. Royal Society of London. Royal Botanic Gardens, Kew. Royal Colonial Institute, London. Royal Society of Edinburgh. Botanic Society, Edinburgh. ' PUBLICATIONS RECEIVED. XX 5 Germany — Bibliothek der Bayer, Akademie der Wissenschaften, Munich. Notgemeinschaft der Deutschen Wissensehaft, Berlin. Senckenbergischen Bibliothek, Frankfurt, A.M. France — Societe Geologique et Mineralogique de Bretagne, Rennes. Societe Botanique de France, Paris. Musee d’Histoire Naturelle, Paris. Societe Scientifique Naturelle, Nantes. Office Scientifique Peches Maritimes. Observations Meteorologique de Mont Blanc. Italy — Societa Agricana d ’Italia, Naples. Instituto di Bologna. Societa Toscana di Scienze Naturali, Pisa. Poland — Univesitatis Liberag Polonas. Spain — Real Academia de Ciencias de Madrid. Real Academia de Ciencias de Barcelona. Academia de Ciencias de Zaragoza. Sweden — Geological Institute, Upsala. Switzerland — Naturforschende Gesellschaft, Basle. Naturforschende Gesellschaft, Zurich. Societe de Physique et d’Historie Naturelle, Geneva. International Labor Office, Geneva. XXII LIST OF MEMBERS. List of Members. Corresponding Members. J Danes, Dr. J. V. David, Professor Sir T. W. E., F.B.S. JDomin, Dr. K. JHedley, C., F.L.S Liversidge, Prof. A., F.R.S. ^Maiden, J. H., I.S.O., F.R.S., F.L.S. jMaitland? A. Gibb, F.G.S Rennie, Professor E. H. JSkeats, Professor E. W. University, Prague, Czecho-Slovakia. The University, Sydney, N.S.W. Czech University, Prague, Bohemia. Assistant Curator, Australian Museum, Sydney, N.S.W. Fieldhead, Coombe Warren, Kingston Hill, Surrey, England. Botanic Gardens, Sydney, N.S.W. Geological Survey Office, Perth, W.A. The University, Adelaide, S.A. The University, Melbourne, Vic. Ordinary Members, Etc. Alexander, W. B., M.A. Appleby, W. E. Bage, Miss F., M.Sc. tBagster, L. S., D.Sc. JtBailey, J. F. . . Ball, L. C., B.E ftBancroft, T. L., M.B. ^Bancroft, Miss M. J., B.Sc. Barker, Miss E., B.A. Barton, E. C., A.M.I.C.E. .. Berney, F. L. -+ Bennett, F., B.Sc. Prickly-pear Laboratory, Sherwood. Sugar Refinery, New Farm. The Women’s College, Kangaroo Point, Brisbane. The University, Brisbane. Botanic Gardens, Adelaide, S.A. Geological Survey Office, Melbourne Street, South Brisbane. Eidsvold, Queensland. Medical School, The University, Sydney. Girls ’ State School, South Brisbane, care of National Bank of Australasia, 4 Queen Victoria Street, London. Barcarolle, via Longreacli. State School, Toowong, Brisbane. t Life Members. | Members who have contributed papers to the Society. LIST OF MEMBERS. Bick, E. W Bradley, H. Burton, M.B., Ch.M. ifBrunnich, J. C., F .I.C. 4 Bryan, W. H., M.Sc. ; Brydon, Mrs. Bundock, C. W., B.A. : Butler-Wood, F., D.D.S. Butler-Wood, Miss I. V., B.D.S. Cameron, W. E., B.A. Cayser, A. A., B.Se. Cilento, B. W., M.D., B.S. . . -IColledge, W. R Colvin, Joseph 4 Cooling, L. E. Cumbrae-Stewart, F. W. S., D.Q.L. Dixon, G. P., C.B.E., M.B., Ch.M. . . JDodd, Alan P Drewitt, G. E. Duhig, J. V., M.B Duhig, Archbishop, D.D. Dunstan, B. . . ^Francis, W. D. Froggatt, J. L., B.Se JGailey, R. Gibson, J. Lockhart, M.D. JGillies, C. D., M.Sc Goddard, Prof. E. J., B.A., D.Sc. . . Graff, R., B.Se xxiii Botanic Gardens, Brisbane. Longueville, North Shore, Sydney. Agricultural Chemist’s Lab., William Street, Brisbane. The University, Brisbane. Department Public Instruction, Bris- bane. 1 1 Kooralbyn, ’ ’ Beaudesert. Permanent Chambers, Adelaide Street, Brisbane. Dornoch Terrace, West End. Tanbun Road, Ipoh, Perak, Federated Malay States. The University, Brisbane. Institute of Tropical Medicine, Towns- ville. Friendly Societies ’ Dispensary, George Street, Brisbane. George Street, Brisbane. Institute of Tropical Medicine, Towns- ville. The University, Brisbane. Wickham Terrace, Brisbane. Prickly- Pear Laboratory, Sherwood, Brisbane. Narrandera, N.S.W. Edmonton, Wickham Ter., Brisbane. 1 ‘ Dara, ’ ’ Brisbane. Geological Survey Office, Melbourne Street, South Brisbane. Botanic Gardens, Brisbane. Department of Agriculture, Brisbane. Courier Buildings, Queen Street, Bris- bane. Wickham Terrace, Brisbane. Club House, The University, Mel- bourne. The University, Brisbane. Medical School, The University, Mel- bourne. t Life Member. J Members who have contributed papers to the Society. XXIV LIST OF MEMBERS. *Grey, Mrs. B. B., F.L.S. . . Greene, Miss A. | Gurney, E. H. Guthrie, F. B., F.I.C. Hamlin, J. C., M.Sc. Hamlin, Mrs. J. C., M.A. }Hamlyn-Harris, R., D.Sc. . . Hard'castle, Mrs. T., B.Sc. . . Hardy, G. H JHawken, Professor R. W., B.A. tHenderson, J. B., F.I.C. Heney, T. W. Hitchcock, L. F. Holdaway, F. G., B.Sc. Hulsen, R. Illidge, T. Jackson, A. G. J Jensen, H. I., B.Sc. t Johnston, Professor T. Harvey, M.A., D.Sc. Just, J. S. JLainbert, C. A. Lloyd, W. JLongman, H. A., F.L.S. JLove, W., M.B., Ch.M Lowson, Professor J. P., M.A., M.B. Ludgate, Miss B., B.A. Lumley Hill, Mrs. E. care of Queensland Trustees Ltd.r Margaret Street, Toowoomba. High School, Wynnum. Agricultural Chemists’ Lab., William, Street, Brisbane, care of Daily Mail, Brisbane. care of Prickly-pear Laboratory,. Sherwood. care of Prickly-pear Laboratory,. Sherwood. Hookworm Campaign, Brisbane. Jinbiggaree, near Dugandan. Biology Department, The University,. Brisbane. The University, Brisbane. Government Analyst, Brisbane. The Telegraph, Brisbane. Priekly-Pear Laboratory, Sherwood. Department of Agriculture and Stock,, Brisbane ( Associate ) . 238 Edward Street, Brisbane. Markwell Street, Toowong. Synchronome Co., Ann Street, Bris- bane. Treasury Chambers, George Street,., Brisbane. The University, Adelaide. care City Electric Light Company,; Boundary Street, Brisbane. care B^nk of N.S.W., Melbourne, Vic. Queensland Correspondence College,, Adelaide Street, Brisbane. Queensland Museum, Brisbane. 1 Wickham Terrace, Brisbane. Lauriston, Wickham Terrace, Brisbane.. Teachers’ Training College, Brisbane. Belle Vue, via Coominya, Esk Line. t Life Members. J Members w7ho have contributed papers to the Society. LIST OF MEMBERS. XXV Marks, Hon. Dr. Marks, A. H., C.B.E., D.S.O., M.D. 1 Marks, E. O., M.D., B.A., B.E. . 4 McCall, T., E.I.C McDonald, S. F., M.D., M.R.C.P. . McMinn, J. . . Massey, Rev. C. H. . . Mathewson, J. H. R., M.B., Ch.B. . , Merrington, Rev. E. N., Ph.D. Morris, L. C., A.M.I.C.E. . . Morton, C., A.T.C.S.M. Moxon, Miss C. Muir, Miss E. Parker, W. R., L.D.S. Parnell, Professor T., M.A. 4 Pearce, Mrs. T. R., M.Sc. . . 4Pound, C. J., F.R.M.S. 4 Priestley, Professor H. J., M.A. . . j 4Beid, J. H 4Richards, Professor H. C., D.Sc. I IRiddell, R, M ! IRoe, R. H., M.A I Saint-Smith, E. C., A.S.T.C. j Sankey, Major J. R. Saunders, G. J., M.Sc., B.E. Shepherd, S. R. L. . . Sirnmonds, J. H., B.Sc. Skertchly, Professor S. B. J. 4Smith, F., B.Sc., F.I.C. t Steel, T 101 Wickham Terrace, Brisbane. Wickham Terrace, Brisbane. 101 Wickham Terrace, Brisbane. Government Analyst’s Department, Brisbane. Fancourt, Wickham Terrace, Brisbane. State School, Kelvin Grove, Brisbane. Murgon. New Farm, Brisbane. Dunedin, N.%. Department of Public Instruction, George Street, Brisbane. Geological Survey Office, Melbourne Street, South Brisbane. Broughton Estate, Toowong (Asso- ciate). Taringa (Associate). 185 Edward Street, Brisbane. The University, Brisbane. “Nattai, ” Beresford Avenue, Chats- wood, Sydney. Bacteriological Institute, Yeerong- pilly. The University, Brisbane. Geological Survey Office, Melbourne Street, South Brisbane. The University, Brisbane. Department Public Instruction, Bris- bane. Figtree Pocket. Geological Survey Office, Melbourne Street, South Brisbane. Flavelle’s, Queen Street, Brisbane. Central Technical College, Ipswich. Geological Survey Office, Brisbane. Department of Agriculture and Stock, Brisbane ( Associate ) . Molendinar, S.C. Line. Hutton ’s Factory, Zillmere. Stephens Street, Pennant Hills, Sydney, N.S.W. t Life Members. 4 Members who have contributed papers to the Society. XXVI LIST OF MEMBERS. Swanwick, K. ff., B.A., LL.B. Swain, E. H. F. Sylow, Paul . . * . The University, Brisbane. Director of Forests, Brisbane. Sugar Befinery, New Farm. ■t Taylor, Hon. W. F Thompson, C. L., B.D.Sc. . . Thynne, Hon. A. J. . . JTiegs, 0. W., D.Sc ?Tryon, H. | Turner, A. J., M.D., F.E.S. i „ Preston House, Queen Street, Brisbane- 89-91 Queen Street, Brisbane. 195 Edward Street, Brisbane. The University, Adelaide, S.A. Department of Agriculture, Brisbane., i 131 Wickham Terrace, Brisbane. Walker, A. B., D.D.S., L.D.S. Walker, Miss Mavis J., M.Sc. tWalkom, A. B., D.Sc. Edward Street, City. The University, Brisbane. Linnean Society House, Elizabeth Bay,. Sydney. Watkins, S. B., M.Sc. Wearne, B. A., B.A. tWeedon, W Central Technical College, Brisbane. . Central Technical College, Brisbane. ‘ 1 Innisfail, ’ ; Wickham Terrace, Bris^ bane. J White, C. T., F.L.S Government Botanist, Botanic- Gardens, Brisbane. tWhitehouse, F. W., B.Sc. . . King 7s- College, Brisbane (Associate) -d t Life Members. 1 t Members who have contributed papers to the Society. PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND FOR 1924. Printed for the Society by ANTHONY JAMES CUMMlNG, Government Printer, Brisbane, 1925. Price: Ten Shillings. PROCEEDINGS OF THE ROYAL SOCIETY QUEENSLAND FOR 1924. VOL. XXXVI. ISSUED 23rd FEBRUARY. 1925. Printed for the Society by ANTHONY JAMES CUMMING, Government Printer, Brisbane. 1925. Price: Ten Shillings. The Royal Society of Queensland. Patron : HIS EXCELLENCY SIR MATTHEW NATHAN, P.C., G.C.M.G. OFFICERS, 1924-1925. President : W. H. BRYAN, M.Sc. Vice-President : . , Dr. E. O. MARKS, B.A., B.E., M.D. Professor R. W. HA WREN, R.A., B.E., A.M.I.C.E. Hon. Treasurer : Hon. Secretary : E. W. BICX. W. D. FRANCIS. '■ \ ■ v ’ . ’ ' . '■ '• Hon. Librarian : Hon. Editor : Professor E. J. GODDARD, B.A., D.Sc. H. A. LONGMAN, F.L.S. Members of Council : J. Y. DUHIG, M.B. Professor H. J. PRIESTLEY, M.A. E. H. GURNEY. Professor H. C. RICHARDS, D.Sc. C. T. WHITE, F.L.S. Trustees : R. H. ROE, M.A. Hon. A. J. THYNNE. Hon. W. F. TAYLOR. Hon. Auditor : Professor H. J. PRIESTLEY M.A. Bankers : QUEENSLAND NATIONAL BANK. LIBRARY of C@NGftE38 f'fiaeiv&D &bb.°H3 CONTENTS. VOLUME XXXVI. Page. No. 1. — Presidential Address : Some Doubts in Queensland Physiography. By E. 0. Marks, M.D., B.A., B.E. Issued 27th June, 1924 . . . . . . . . . . . . 1-18 No. 2. — A Suggestion for a Biological Laboratory on Stradbroke Island for the Protection of Ceratodus. By T. L. Bancroft, M.B. Issued 27th June. 1924 . . . . . . 19-20 No. 3. — The Development of Buttresses in Queensland Trees. By W . D. Francis, Assistant Government Botanist. Plates I. -VI. Text figures 1-7. Issued 20th August, 1924 . . 21-37 No. 4. — On the Synonymy of Australian Flies belonging to the genus Actina (StratiomyiiDuE). By G. H. Hardy. Issued 20th August, 1924 . . . . . . . . . . 38-40 No. 5. — On a New Species of Melaleuca (Family Myrtaceas) from Southern Queensland. By E. Cheel and O. T. White. Text figure 1. Issued 20th August, 1924 .. .. 41-43 No. 6.— The Geology of the Silverwood -Lucky Valley Area. By Professor H. C. Richards, D.Sc., and W. H. Bryan, M.Sc. Plates VII. -XX. Issued 20th August, 1924 . . . . 44-108 No. 7. — New Queensland Loricates. By A. F. Basset Hull. Plate XXI. Issued 1st September, 1924 . . . . 109-116 No. 8.- — The Artesian Waters of Queensland. By P. C. Tibbits , A.M.I.E. Plates XXII. -XXIV. Issued 23rd February, 1925 117-128 No. 9. — On a New Species of Pandanus from North Queensland. By Count Professor U. Martelli, (Florence, Italy). Plate XXV. Issued 23rd February, 1925 . . . . . . 129-130 No. 10. — Radiolarian Jaspers in the Brisbane Schist Series. By Professor H. C. Richards, D.Sc., and W . H. Bryan, M.Sc., . University of Queensland. Plate XXVI. Issued 23rd February, 1925 . . . . . . . . . . . . . . 131-135 No. 11. — A New Rotifer of the Melicertan Family. By W. R. Colledge. Plate XXVII. Issued 23rd February, 1925 137-138 No. 12. — Geological Notes on the Country between Childers and Mundubbera. By H. I. Jensen, D.Sc. Issued 23rd February, 1925 139-144 Abstract of Proceedings List of Library Exchanges List of Members ' ■■ * L.t •*«*** * ■■’v.rusc l \ * * ; ^ v. xix. xxi. Vol. XXXVI., No. 1. Proceedings of the Royal Society of Queensland for 1924. Presidential Address. By E. 0. Marks, M.D., B.A., B.E. ( Delivered before the Royal Society of Queensland , 31st March , 1924.) It is a source of much gratification that it has not fallen to my lot to record the departure from us of any member on that great voyage of discovery whence- there is no return. It is, however, with great regret that reference must be made to the passing of a former esteemed and active member of this Society and well-known figure in scientific circles in Brisbane: Charles James Wild. He was for many years on the staff of the Queensland Museum, his chief work being in the realms of entomology, conchology, and the mosses. For five years he was Acting-Director of the Museum, retiring from that position and from active scientific work in 1911. Of late years he had lived in Dalby, where he died, and in consequence he was little known to the younger generation, but those of us who had occasion to seek his help at the Museum will long remember the ready kindness with which he gave it. The past year has been notable for completing the first century of white occupation of what is now Queensland, and our thoughts natur- ally turn to those far off events that were the dawn of our history and to the founders of our State. It is not easy for us nowadays to realise the difficulties and dangers which had to be overcome and the hardships and uncertainties which had to be endured by those early settlers, both official and civilian, to whom we owe the opening up of the hitherto unknown land. Of the explorers who risked, and not infrequently lost, their lives in the investigation of the new country, of the early scientists whose enthusiasm was rewarded by new and interesting discoveries at every turn, too much cannot be said in their honour, but to me it has always seemed that the real heroes have been those stout-hearted pioneers, men and women, who, regardless of personal risk, hardship and isolation, ventured far out into the “bush” to live and to find by bitter experience the limitations as well as the capabilities of the land of their adoption. It is to these bold colonists, courageous experi- menters, that we owe largely the advance of the last century and our own position in the land. Long may they be honoured in the country which they served so well! R.S. — B. 2 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. A hundred years ago “scrub,” forest and plain, with their inhabitants, man, bird and beast, were existing in a sort of natural balance of forces arrived at after probably many thousands of years of similar conditions and isolation from invaders. The advent of civilised man with his implements of construction and destruction, with his useful and harmful animals and plants, has introduced forms of life which have run riot in their new home and has caused the diminution or extinction of some of the native forms. To the student of nature the effects of this hundred years’ invasion must be a frequent source of contemplation, many aspects of which have an important bearing on our future welfare. Apart from the problems afforded by the introduced pests, are there not such questions as the influence, if any. of tree destruction on rainfall, the effect of continued grazing on the natural grasses and on the carrying capacity of the land, the dense growth of trees in country formerly open forest, and many more problems the investigation of which now would be of value in guiding the future development of the land. The study of the hundred years’ invasion has economic as well as a scientific and sentimental interest. To many scientific workers in this State probably the most interesting event of the year has been the Pan-Pacific Science Congress held in Sydney and Melbourne in August and September. Besides the actual scientific papers and exchange of views on matters of common concern to countries of the Pacific littoral, the Congress afforded opportunity for making that personal acquaintance with overseas workers which is so much to be desired, since it helps to counteract the insularity which scientists, in common with the rest of humanity, are liable to develop. Some of the visitors wre had the pleasure of meeting in Brisbane. Professor J. Cossar Ewart, F.R.S., of Edinburgh University, celebrated for his experiments in animal-breeding, very kindly gave us two most interesting lectures: — “The History of Feathers,” and “The Scientific Breeding of Wool.” The latter lecture was held in conjunction with the United Pastoralists’ Association, and it was very gratifying to see present so many representatives of our chief industry. ****** The slaughter of a beautiful theory by a brutal fact has always afforded me as unholy a gratification as it ever can have given the distinguished author of the expression. Looking back the short space of twent}r-four years since first attending University lectures on scientific subjects, various theories and assumptions then taught and accepted as practically established facts have since been found to be erroneous or to require great modi- fication. When such an apparently justifiable conclusion as the gradual cooling down of the earth lias been thrown in doubt by the advent of radium, when our former atoms or indivisibles are now a PRESIDENTIAL ADDRESS. 3: complex of electrons, when even Newton’s laws appear to be somewhat warped, not to mention various minor but firmly established ideas that have ‘'gone west,” it is not strange that a natural scepticism has become more pronounced. In regard to one Queensland subject, Physiography, I have long been an unbeliever of much that has been written on it, and particularly of the methods employed by most of the writers in arriving at their conclusions. In the hopes that an expression of these doubts may not be without utility in the further study of a very interesting subject I venture to bring to your notice : Some Doubts in Queensland Physiography. That branch of Geology known as Physiography provides an excep- tionally fascinating study. It is a never-failing source of interest wherever one travels, constantly presenting features for inquiry and problems for solution. It is a science of observation and deduction, but in its very nature is neither an experimental nor an exact science. Its more advanced deductive development is largely owing to American geologists, but the main fundamentals are probably not due to any one nation more than another, and are well established in the older works. In its more modern form Physiography is somewhat richly endowed with a polyglot terminology of its own and is the result of much a priori reasoning based on a necessarily incomplete knowledge of natural pro- cesses. It provides a field for the exercise of the imagination such as no other branch of Geology can do, and offers facilities for this unequalled by the less picturesque branches of the science. For instance, in a few weeks’ journey, a physiographer can tour through our State and see something of our western plains, the coastal ranges and lowlands, the northern tablelands and gorges, and the coastal scenery. He may then retire into seclusion with a map and write learnedly of uplifts and downthrows, horsts and “ senkungf elder, ” fold mountains and rift valleys, plateaus and peneplains, continental divides and continental shelves, rivers and river captures, and so on, with little fear that any one can disprove his thesis. It is little wonder that modern Physiography has had so many devotees when it offers such alluring fields for conquest: none of the tedious mapping of boundaries, no slow puzzling out of geological rela- tionships, no close and difficult investigation of petrological and palaeon- tological problems. While he that runs may read, the physiographer in a railway train scarcely even needs to read ; the scenery tells its own tale. When one allows for a large personal equation, it is not altogether surprising under these circumstances that physiographers may produce widely differing interpretations. It is surprising that such an event does not appear to shake their faith in the validity of their own conclusions. To the present writer it seems to indicate that the foundations on which 4 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. their reasoning is based must be defective, as otherwise they would all surely arrive at the same general conclusion, by whatever route might be adopted. Owing to the commendable activities of the Great Barrier Reef Com- mittee, on the inauguration of which the Royal Geographical Society of Australasia, Queensland, is greatly to be congratulated, much attention is being directed at the present time to the Reef and its many problems. Probably the most interesting and controversial of these problems is its mode of genesis, and this is inseparably connected with the recent geo- logical history of the continental shelf, the coast line, and the back country. For a study of the origin of the Reef it is essential that there must be brought to it all such knowledge as can be derived from the physiography of Eastern Queensland. After a period of comparative stagnation there is thus a prospect of fresh physiographical work in this country. With this prospect in view the writer has endeavoured to ascertain what is the present position of our knowledge of Queensland Physiography, and especially the basis and reliability of the deductions that have been made. The result, on the whole, has been very unsatis- factory, facts being scarce but fancies plentiful. In the following pages some criticism has been made of the individual writings on the subject in order to show the reasons for this dissatisfac- tion and how unreliable many of the conclusions or assumptions have been. Unless a conclusion can be thoroughly relied on, it is useless or worse than useless for further deduction, no matter how interesting or plausible it may be. Our one object must be to obtain positive know- ledge, free from conjecture, if we are to progress at all in solving the problem of coral reefs or any other problem connected with Physio- graphy. We cannot be content with probabilities, for one probability dependent on another becomes usually an improbability. How few workers seem to remember that even a probability so high as 2 to 1, based on another probability of 2 to 1, is really an improbability of 5 to 4. The criticism is made in many cases of the work of friends, and in every case in the friendliest spirit. It is with the general method of applying physiographic principles that fault is found, and not with the individual writers. Of more or less general interpretations of the Physiography of Queensland we have three : by E. C. Andrews, Griffith Taylor, and J. V. Danes. There are also references to the geotectonic structure by Sir T. W. E. David, and references to more local conditions and to the coast by C. Hedley, Jensen, Wearne and Woolnough, Poole, Richards, W. M. Davis, and Jardine, as well as the very valuable descriptions and refer- ences contained in the various reports of the Geological Survey. On the whole, however, the officers of that Survey have shown a disinclination to do more than describe the physical features in the areas dealt with. PRESIDENTIAL ADDRESS. 5 Few teachers of Geology can have been the object of such a personal affection and admiration on the part of their students as has Professor Sir T. W. E. David, F.R.S. The writer has met many of his old students and knows that by them Sydney University could scarcely be mentioned without at the same time mentioning ‘'Prof. David.” He is regarded generally as the foremost geologist in Australia, and rightly so. It is therefore in no carping spirit that criticism is. made of some aspects of his writings on Queensland, but rather in the hope that having criticised Professor David no one else can feel offence if criticised too. It has been felt too that, owing to the great authority which his statements carry, the basis of those statements requires particularly careful investigation. In his Presidential Address to the Royal Society of New South Wales in 1911 he took as his subject, “Notes on Some of the Chief Tectonic Lines of Australia,” and dealt chiefly with the structural Geology, the Physiography forming a minor feature, and he dealt more with recorded observations than deductions. In places, however, where observations are wanting, he appears to fall into the lure of making assumptions without the facts to support them and does not always clearly differentiate between what is observed fact and what is little more than pure supposition. For example, in dealing with Queensland, he commences by saying, “Reference to Plates 1 and 2 of this address reveal the following dominant features : — (1) Ranges mostly of palasozoic rocks forming the highlands of the main divide, an ancient peneplain trending about N. 30° W. . . . ” Ranges and highlands presumably do not form a peneplain at the present time. Plates 1 and 2 (a photograph of a. relief model and a map) do not indicate a present peneplain and there seems to be no evidence — certainly none is given — that there ever was any more of a peneplain than there is at the present time. As we shall see. later, nearly every other writer assumes this ancient peneplain. It may or may not have existed, but since it forms the base on which much of the subsequent theories have been built up it obviously demands the very closest scrutiny. On page 45, Professor David says, “Trend lines are shown by the belt of serpentine to the west of Gympie and the immense mass of serpentine to the north of Rockhampton. The latter trends N.W. and S.E. true and is evidently situated on a zone of heavy fractures. These extend all the way from Gladstone to Herbert Creek at Broadsound. The trachyte volcanic centres at Yeppoon and Berserker ranges are close to these major lines of fracture. They are doubtless part of the group of great fractures along which the eastern side of the divide has been stepped down below sea level. There is both physiographic and strati- graphic evidence for this fault at Curtis Island.” Here Professor David makes the following four assumptions: — (1) A zone of heavy fractures; (2) he then assumes them to extend from Gladstone to Broadsound, and more or less suggests a causal connection between the trachyte volcanic centres and the assumed fractures, which 6 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. are then assumed (3) to be part of a group of fractures along which the eastern side of the divide is assumed (4) to have been stepped down below sea level. It comes as a relief to be informed that there is both physiographical and stratigraphical evidence for this fault at Curtis Island, though the evidence is not specified. It would be interesting to estimate the probability of each of the above assumptions and then calculate the improbability of the sequence. The next paragraph deals with a fault that is presumably part of the above zone of faulting, since it coincides in portion of its course. In a map illustrating a report by Mr. L. C. Ball there had been indicated a small area of Trias-Jura strata. The western limit of this, as sketched on the map, is straight, and coincides with the river Boyne. From this one small indication on the map, to which there is no reference in the text, and perhaps influenced by the straight river valley, Professor David tells us : — ‘ ‘ The Boyne River follows this line of fault which presumably throw's E. 40° N. The channel between Curtis Island and the mainland is on a continuation of this fracture. Beyond Keppel Bay it seems to divide, an eastern branch going to Shoalwater Bay and the Northumber- land Islands, the western to the estuary of Herbert Creek and Broadsound.” The small area of Trias-Jura strata was only indicated on the map and had not at that time been geologically surveyed. Work which has been done since shows it to be of a later age and that there is probably faulting, but the direction and magnitude of this are not stated. Evi- dently more work requires to be done before a conclusion can be arrived at. Professor David, on the strength of the sketch mapping, without description in the text, deduces a fault 150 miles long. It is interesting to note that Mr. Dunstan, the Chief Government Geologist, makes no reference to this supposed zone of faulting when dealing wfith the geological history of Keppel Bay, even though he refers to the Fitzroy River having at one time passed between Curtis Island and the mainland. He apparently did not appreciate the physiographical and stratigraphical evidence of a fault at Curtis Island to which Professor David refers. On page 46, Professor David says, “The Cumberland and Whit- sunday Islands are almost certainly horsts amongst a network of faults. ’ ’ He gives no evidence for this, apparently assuming both the existence of the faults and that the islands are horsts. Why they should not be merely the higher points of a sunken land surface is not mentioned. This simpler view has recently been adopted by Hedley and Richards. So far as the writer is aware, all physiographers who have visited the east coast of Queensland, from J. B. Jukes to W. M. Davis, are agreed that it is a sunken coast line. This excludes, of course, the minor varia- tions in level indicated by raised beaches and river terraces. This general sinking of the coast line appears to be the one physiographical deduction, which may be regarded as well established by field evidence PRESIDENTIAL ADDRESS. 7 (provided by the islands and inlets), though there is much yet to be learnt about it and about the influence of rock structure on the coastal forms. This influence may take effect directly by the varying resistance to wave action and indirectly by determining the land form prior to sinking, whereby a sunken surface of low relief would produce a coast of low relief. Equally ns great care is required in the coast study as inland, and it appears to be equally as difficult to avoid jumping at conclusions, for we find that even our usually cautious friend Professor Richards says, with Mr. C. Hedley, in reference to Torres Strait: — “The region is one of considerable relief and has apparently resulted from a series of fractures along two lines at right angles to one another. It is highly probable that there is a causal connection between these fractures and the Tertiary fold mountains of Papua.” One cannot help wondering whether the sinking of an area of high relief, the relief due to denudation modified by rock structure, could not account for the present appearances without any series of fractures. When the fractures have been observed in natural sections let us believe in them, and even then their direct responsibility for the land form still requires to be demonstrated, for the fractures might be ancient ones and only indirectly responsible, through denudation, for the present surface relief. It is notorious that appear- ances are often deceptive. Physiographic appearances are particularly liable to deceive, and we cannot be too careful. Furthermore we know how rare it is for faults demonstrated in cuttings and mines to directly affect the configuration of the land other than by differential weathering, and it should be just as rare on the coast line. Should observations show that these doubts are groundless and that the fractures do exist and are responsible for the surface relief, the probability of a causal connection with the Papuan fold mountains is no greater than the probability that there was no connection whatever. We need more evidence than mere geographical proximity. The evidence may be obtainable. Some writers regard Torres Strait as a rift valley. The evidence for this seems to be nil, unless we are to regard any shallow depression below sea level as a rift valley. Of Australian physiographers, probably no name is better known that that of Mr. E. C. Andrews, the Chief of the Geological Survey of New South Wales. His writings are necessarily mostly concerned with that State, but sometimes include Queensland within their scope. In his “Geographical Unity of Eastern Australia in Late and Post-Tertiary Times,” the main deductions are based on observations in the South, but are made to apply to Queensland, which constitutes more than half of Eastern Australia, and reference is necessarily made to such Queensland observations as have been recorded. Deductions arrived at from a consideration of the mountain masses in the South are considered by him to apply to the whole of the eastern highlands as far as Cape York. In his reference to Queensland it is 8 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. difficult to differentiate Mr. Andrews’s statements of facts from his deductions and assumptions, particularly in regard to faulting and warping. Thus he gives as an appendix a list of some important fault scarps and senkungsf elder in Eastern Australia, presumably regarding them as established facts. In the Queensland examples, are (2) Bellenden Ker (an uplifted block) and (6) North Queensland scarp (uplift) ; Professor David regarded the coast as having been faulted down, leaving the highlands hanging, surely not quite the same thing; (10) Darling Downs faults: these were inferred by Messrs. Wearne and Woolnough on physiographic grounds and have lately been questioned by Mr. J. H. Reid; (11) Mount Elliot, near Townsville, inferred by Mr. W. G. Poole on physiographic grounds; (20) the MacPherson Range, inferred by Mr. Andrews but since disproved by field work on the Queensland side. Not one of these alleged Queensland faults has a basis of unques- tionable evidence, for none of them has been confirmed by detailed examination of the ground or observed in any section, and in none of them has the possibility of differential weathering been excluded. Grave doubt or actual disproof has been put on some and different interpreta- tions put on others. In maintaining the existence of a late Tertiary peneplain, Mr. Andrews places great importance on the “deep leads” of Victoria and New South Wales, which have yielded a fossil flora as far north as the Darling Downs. He mentions (p. 449) that the North Queensland leads have not yielded the characteristic flora, and then on the strength of the fossil evidence he says that “a mild and uniform climate must have extended right from Hobart to North Queensland in Upper Pliocene time. Incidentally this would imply a non-mountainous country.” Mr. Andrews states that the uniformity of flora, coupled with the fact that the leads occur both on the eastern and western fall of the highlands, points irresistibly to the conclusion that the eastern side of Australia in Date Tertiary times consisted of a slightly roughened surface not raised much above sea level. This may have been true, but the conclusion seems far from irresistible if we consider the present-day distribution of flora in Queensland, while according to Mr. Andrews himself it is without any fossil evidence to support it in the greater portion of Eastern Australia. Mr. Andrews expresses astonishment that neither Murray, Baron von Muller, nor Sir F. McCoy perceived this physiographic “fact.” If this is regarded as a fact, one cannot be surprised at a difficulty in distin- guishing facts from theories. There occur in various parts of Queensland, as for instance near Brisbane, Gladstone, Duaringa, Boyne Valley, besides the Darling Downs, deposits containing dicotyledonous leaves vaguely referred to as of Tertiary age. Near Brisbane and at Gladstone they occur at sea level, and at Duaringa are at a low level. Near Brisbane they are covered in places by basalt and may be of the same age as those of the Darling Downs. If these are of the same age as the “deep leads” they would PRESIDENTIAL ADDRESS. 9 certainly destroy tlie theory of an uplift at the close of the Tertiary period. A very great deal of field geology and palaeontological work is required to be done on them before any reliable deduction can be made from their mode of occurrence. The same may be said of the volcanic rocks. We have in Queensland a great variety of Tertiary and Post-Tertiarv lavas of varying age. Some, as in the MacPherson Range, now form mountains up to 4,000 feet in height and owe their present contrast in elevation with the sur- rounding country to the more rapid denudation of the softer sedimen- taries. Others, as for instance near Brisbane and in North Queensland, occur in the existing valleys. The lava flood on the Copperfield River passes through a narrow gorge, leaving a rim of basalt like a highwater mark on the sides, only about 15 feet above the present granite bed of the river, so recent has been the flow. It will be understood therefore what very great caution is required before coming to any conclusions from the association of basalt with anything in the nature of a lead, but we have Mr. Andrews’s definite summary : — (1) The Tertiary peneplain extended from north to south across the continent. The flora was uniform and evidenced a mild to tropical climate. (2) The leads of the Pliocene or of the “Newer Yolcanics” are to be found all along the Eastern continent. They all evi- dence similar floras, and all evidence subsidence accompanied by filling of the channels by similar continental deposits and final burial under floods of basaltic lavas. (3) The basalts of the “Newer Yolcanics” are similar in appear- ance along the whole eastern side of Australia. (4) The Kosciusko period is also noted for its production of similar topographic features extending from north to south of Australia. So also are its rejuvenation phases. Mr. Andrews’s interpretation may or may not be the correct one, but with the information at present available it can hardly be regarded as anything more than conjecture so far as it applies to Queensland. Profesor Griffith Taylor, in his account of the physiography of Eastern Australia, refers at some length to the North Queensland rivers, the Burdekin System, the Fitzroy System, and the South Queensland rivers. Prom peculiarities of their courses, as depicted on the maps, “crossed forks” and “boathook bends,” he deduces river captures and breached divides. He considered that formerly the divide was much further east, coinciding more or less with the areas of granite intrusion, and that the Burdekin and Fitzroy as well as other streams had captured the headwaters of the western-flowing rivers. One cannot help noticing that, with the exception of the basalt flows, he appears to neglect alto- gether the influence of geological structure on the courses of the streams. 10 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. and to pay little attention to land forms except in so far as they immedi- ately affect the rivers. In a general way he appears to agree with most of the other writers that in Tertiary times Eastern Australia attained a very mature condition, though he does not refer to it as a peneplain. That the shorter and more active eastern-flowing streams of the present day are encroaching on the headwaters of the western-flowing streams and that the divide is consequently in process of shifting west- wards seems a reasonable inference. How long this has been going- on and to what extent modified by earth movements are problems to be solved, in which Professor Griffith Taylor’s methods will form an important but not the only consideration. Mr. C. Hedley, well known for his studies of the marine fauna and the continental shelf, has been impressed by the fact that where the continental shelf is widest the divide is also farthest from the coast. This applies particularly to the region near the tropic line, where the largest streams, the Bur dekin and Fitzroy, have such remarkable courses. According to Mr. Hedley ’s theory — which he does not put forward as anything more than a theory — the old peneplain had a radial drainage ; that is to say, the rivers drained more or less from the centre of the land towards the coast. Pressure from the Pacific dropped the Thomson and Carpenter deeps and raised the present margin of the continent, giving rise to the present “marginal” drainage. He considers that the wide continental shelf represents a buttress area which protected the region behind it, and that the less rapid or smaller movement in the area sheltered by the buttress was not greater than the streams could cope with. Consequently these streams lowered their beds as quickly as the coast country was raised, and still occupy more or less their original positions. They are thus relics of the old radial drainage. It is some- what curious that Mr. Hedley supposes a thrust from the Pacific, whereas Mr. Andrews regarded the thrust as coming from the opposite direction. The fact that the upper parts of the Burdekin and Fitzroy are in wide valleys and undulating country and that in their lower portions both rivers cut through a range of hilly country in a narrow gorge has appealed strongly to Mr. Hedley, as it has to other writers. The geo- logical structure is not considered in reference to its possible effect on the topography. Dr. J. Y. Danes has made some very valuable observations on the river systems, particularly of the exitless drainage basins of Lakes Galilee and Buchanan, to whose importance in the study of Queensland Physiography he drew attention. His views are of especial value, for he spent a strenuous eight months in field work, much of it far from the beaten tracks. In his writings he clearly distinguishes between observations of fact and theoretical deductions therefrom. In the main he agrees with Mr. Andrews’s interpretation, but Mr. Andrews does not refer much to river drainage, and this has been Dr. Danes’ chief study. In his inter- pretation of the river systems he disagrees with both Griffith Taylor and PRESIDENTIAL ADDRESS. 11 Hedley. He sees in many of the present rivers the remains of separate drainage basins, such as Lakes Galilee and Buchanan are at the present day, and which lay in warpings of the old peneplain. According to this view the eastern- and western-flowing streams, formed since the elevation of the peneplain, have captured these basins, and so formed the compound .streams. The influence of geological structure is scarcely referred to. Mr. W. Poole, in 1909, drew attention to certain features in North Queensland indicating probable capture of portion of the Mitchell River headwaters by the Barron, uplift of the present highlands, and depres- sion of the coast. He also refers to the Burdekin River, especially the gorge and falls in the lower course, which he, like other writers, regarded as a sign of rejuvenation in comparison with the more mature forms exhibited further up-stream. In a paper on the Geology of the West Moreton District, Messrs. R. A. Wearne and W. G. Woolnough, in 1911, dealt with the physio- graphical features of the Main Range, the contours of which they thought showed two successive uplifts. They considered that in Cainozoic time the divide was further to the east and the district a peneplain. This was uplifted 2,000 feet and subsequently another 2,700 feet in late Cainozoic time. Extensive trough faulting produced the Lockyer and Fassifern troughs, thus accounting for the escarpment of the Main Range and for the mature appearance of Fassifern and Lockyer downfaulted blocks. Professor Richards has expressed his agreement with the above deductions of block faulting which had previously been suggested by Dr. H. I. Jensen. Recently Mr. J. II. Reid has examined much of the area and finds himself in disagreement as to the block faulting. After a close examina- tion of the Main Range he could find no stratigraphical evidence of extensive faulting or disturbance of the strata, while a consideration of the levels at which the basalts occur was entirely in its disfavour in those regions to the north of Mount Castle. He says, ‘ ‘ The evidence disclosed by field work is wholly against block faulting having taken place and the origin of the Main Range dividing the Moreton District from the Darling Downs must be attributed to erosion as the main factor, even after a con- sideration of all the evidence which has been adduced previously in favour of block faulting.” As Mr. Reid very rightly points out, if the Main Range were a fault scarp the downthrow side should be basalt as well as the upthrow, but it is not. Dr. H. I. Jensen has dealt in many of his papers with the physio- graphy of the regions under his observation. In his account of the volcanic area of East Moreton he gives an excellent description of the topography. He also records his views of the physiographical history of the region, and in forming his fairly definite conclusions he places importance on what he terms the Woodford peneplain. This is an area of deeply weathered granite with a gently undulating surface about 500 feet above sea level, and is regarded by him as peneplain of Tertiary age. 12 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Between Woodford and Brisbane there exist several other areas of granite or granodiorite, and a study of these has an obvious importance in considering the Woodford area. They occur as intrusions in the more or less metamorphic rocks we know as the Brisbane Schists. The Samford Granite area is some 6 miles east and west by four north and south, and is formed of a deeply weathered granodiorite. Its surface forms what would, if larger, have been called a peneplain, being a gently undulating area with broad shallow valleys similar to that at Woodford. Completely surrounding it, however, the schists form steep hilly country with narrow valleys. To the west Mounts Nebo and O’Reilly are over 2,000 feet in height, and everywhere round the margin of the granite., which in places extends high up the sides of the hills, the summits are schist. Except where the South Pine River makes its exit by a narrow valley through the schist country the Samford area forms a completely enclosed basin. The surrounding hilly schist country is far from being a peneplain, and it seems impossible to avoid the conclusion that the apparent contrast in maturity is really an expression of the weathering character of the two-rock types and nothing whatever to do with the relative denudational age. In the light of the Samford miniature pene- plain, what are we to think of the similar but more extensive occurrence at Woodford? This aspect of the question has left the writer wondering as to the extent to which simple denudation might account for the features described by Dr. Jensen. In a suggestive paper read before this Society in 1922 on “Some Geological Features of Northern Australia,” Dr. Jensen deals with the structural geology. Unlike other writers, he regards the Eastern Cordil- lera belt as extending northwards only to the vicinity of Mackay, and forming a geological unit distinct from the North Australian massif and the rocks folded upon it, which together form the remainder of the eastern highlands to the north. In 1912, before this Society, the present writer gave an account of the Burdekin Falls and Gorge, together with the country for about 100 miles upstream from the falls, this including the features which had been the subject of comment by previous writers. The observations went to show that the contrast between the apparently mature country upstream and the gorge and falls was essentially due to the different weathering of the granite and felsite rocks which form the two very distinct types of country. Attention was also directed to the importance of ascertaining the gradient of streams as an aid to estimating the maturity, for the apparently mature part of the Burdekin under review had a fall of 4 feet per mile, which is greater than what would be expected in a mature stream, while that part of the Fitzroy System from Boolburra to the sea lias a fall of less than 1 foot to the mile, although it includes the supposed rejuvenated portion. PRESIDENTIAL ADDRESS. 13 Since writing this address I have been privileged, through the cour- tesy of Dr. J. P. Thomson, Hon. Secretary of the Royal Geographical Society of Australasia, Queensland, to see the proofs of a paper on the ‘ k Physiography of the Lower Fitzroy Basin,” by F. Jardine. Mr. Jardine, in an excellently prepared article, gives a careful description of the physical features in the area he deals with. He considers most of the features are the result of differential denudation of an old plateau surface. Unfortunately he does not rely much on his own impressions, but prefers to accept without question the views of Professors David and Taylor. On page 36 he says, ‘ ‘ Physiographieally there is little evidence in this area indicating a fault shore line, as it undoubtedly is,” and quotes Professor David as the authority for this faulting. The present writer has already criticised this assumed faulting, and Mr. Jardine ’s trust in Professor David’s assumptions justifies the criticism. There is one misquotation which it is desirable to point out. Mr. Jardine says, “ Lionel Ball shows (quoted by Professor David) that there is a basin of Burrum Beds (Trias or Trias- Jura) thrown down against Devonian rocks. ’ ’ As mentioned elsewhere, Mr. Ball did not show or refer to this fault; it was an assumption of Professor David’s. In regard to the evolution of the Fitzroy System, Mr. Jardine appears to follow without question Professor Griffith Taylor’s theory of a breached divide. Unfor- tunately he does not refer to the rival theories of Hedley and Danes, nor does he state why simple differential denudation might not account for the whole of the features, including the “Gap.” It is to be hoped that Mr. Jardine will continue his investigations of this important part of the river and give us some idea of the length of the course through the highland mass, and the character of the stream and rocks upstream therefrom. On page 18 Mr. Jardine remarks that the Fitzroy, occupying a broad valley, differs from most of the rivers of Eastern Australia which occupy canyons. In Queensland these canyon streams are surely the exception, while such important rivers as the Burdekin, Burnett, and Brisbane show many striking resemblances to the Fitzroy. The present writer does not desire this criticism to detract in any way from the very real value of Mr. Jardine ’s excellent work, but hopes that it will be continued. In the endeavour to obtain a general view of the physiographical history of Queensland as expressed by the various writers on the subject, it is found that all are agreed as to the sunken eastern coast line. The evidence of this foundering seems clear and consistent, apart from minor variations in level, and is provided by the inlets and islands, and to some extent by the soundings. One finds a remarkable divergence of views concerning the history of river drainage divides and so on, a divergence which means that two out of three views must be wrong and not improbably the third. 14 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. There is a very general consensus of opinion that the surface of Queensland was a peneplain in late Tertiary times, though doubtless there are exceptions, as for instance, Mr. J. H. Reid, who considered that peneplanation had not been attained in Tertiary times in Western Moreton district. On the very existence of this peneplain the subse- quent reasoning of many of the writers depends. So far as any evidence is offered, this late Tertiary peneplain appears to be almost pure supposi- tion. It may or may not have existed; far be it from the writer to express any opinion ; but, being an absolute essential to the theories, he feels it demands the closest investigation and some really strong evidence before it can be accepted. At present it is merely a physiographic myth, the least disturbance in which must destroy the card-house theories built on it. Until it is proved, the theories depending on it are well nigh valueless. There is another feature common to most of the writings, and this is the neglect of geological structure and its influence on land forms and drainage. It is to this aspect, so strongly emphasised in such works as Sir A. Geikie’s '‘Scenery of Scotland,.” that the writer desires to draw especial attention, for it is of importance in the interpretation of land forms in Queensland, as elsewhere. As a piece of pure reasoning concerning the results of denudation under assumed idealised conditions, Professor W. M. Davis’s “Cycle of Erosion ’ ’ forms an exceedingly useful and illuminating help in the study of land surfaces. It is not likely that its author ever intended it to be applied indiscriminately, regardless of rock structures and river fall. He certainly appreciated the effect of the varying resistance of rocks to weathering, which would enable one stream to develop a mature surface while another might yet be in a Y-shaped valley. There is another factor in rock weathering which requires to be considered besides mere resistance to weathering, and this is the ratio existing in any rock between its chemical and its physical hardness ; that is to say, the ratio between the rate of softening of the general surface by chemical action and the rate at which the stream can cut down its bed by mechanical action. This ratio for any one rock is necessarily variable with the climate and with the power and fall of the stream, as well as with the rock jointing, but a consideration will show that some rocks could only form gorges and juvenile forms under extreme con- ditions. For instance, a solid basalt is very tough and resistant to mechanical agencies but is not chemically particularly resistant. We are familiar with basalt-capped mountains comparatively flat on top but edged by precipices, which are due to the weathering away of the softer rocks beneath, and over the edge of which the streams form waterfalls, into the gorges below. Yet what a short distance above the fall on the top of the basalt does the stream form a Y-shaped valley, even under such extreme conditions as to declivity. If we consider some of the coarse-grained granitoid rocks which undergo decomposition to a con- siderable depth and yet are, when fresh, mechanically very hard, it seems PRESIDENTIAL ADDRESS. 15 likely that only under exceptional conditions could they be capable of producing any other than comparatively mature forms. This form is characteristic of extensive areas of granitoid rocks in Queensland. On the other hand there are rocks, such as many sandstones, which are chemically resistant but are physically soft, and these would readily be cut into gorges where the general elevation permits. Most rocks occupy an intermediate position, but the ratio between their chemical and physical weathering must have a profound influence on the land form quite apart from the total resistance to weathering. As an example of this principle and the necessity for its consideration, one might refer to the upper Flinders River. The existence of gorges and the changes in direction have led to some inferences as to the capture of the head- waters of the Thomson by the lower Flinders. Now the uppermost part of the Flinders is in more or less undulating wide-valleyed granite country. Lower down, the river is in a narrow gorge cut in sandstone, which in many places is covered by a basalt flow. Lower down again, when on the soft calcareous shales of the Rolling Downs series near Hughenden, it is in gently undulating country. The three conditions of the river coincide with rock changes. The Gilbert River, draining the northern side of the same tableland, is also, when in granitoid rocks, comparatively mature, but gets into gorgey country where the granitoid rocks are covered by sandstone. Surely these changes must be regarded as due to the changes in rock structure with which they coincide, rather than to a rejuvenation which may or may not have taken place. While it seems difficult to avoid the conclusion that gorges and V-shaped valleys are necessarily a sign of youth, what the writer wishes to emphasise is that undulating country with wide valleys is not neces- sarily a sign of maturity. The rock structure and river fall must also be taken into consideration. When we regard Queensland as a whole, the mountainous country — the so-called cordillera — is mainly composed of altered ancient sediments and igneous rocks, while the plains are mainly composed of softer sedi- mentary rocks. Though of smaller extent, the coastal plains, contrasted with the coastal ranges, usually show the same rock contrast as do the western plains. Is it not possible that differential weathering is in the main the direct and essential cause for these topographical features, whether on a small scale or over wide areas ? The earth movements which brought the older rocks up to ti*eir present relative position may be thus merely the remote and indirect cause of the present surface configuration, all traces of the movement having perhaps long since been removed by denudation. In any case differential denudation, whether locally or over wide areas, must be a very important if not predominant factor in determining land forms. To the writer it seems futile, therefore, to make physiographieal deduc- tions about any area whose geological structure is not known in sufficient detail, and without considering differential weathering. 16 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. As the Barrier Reef investigations are certain to give rise to renewed interest in and further study of Queensland Physiography, the writer would like to conclude with an appeal to those about to take up the work. It is an appeal to treat Physiography as they would any other branch of science, to require for it as firm a basis of carefully collected and cautiously co-ordinated fact. There are things we know and things we do not know, and probabilities belong to the latter class. They are obviously dangerous as foundations for subsequent reasoning and should be avoided. Assumptions are even more dangerous, for they are more insidious. They are both utterly destructive of the reliability of any resulting conclusion, and reliability is surely the one essential require- ment of science. Let us demand for Physiography the same amount of positive evidence, the same diligent search for supporting facts, the same care in examination of possible negative evidence, the same detailed study and observation as would be required in any other subject. Should the information obtained or obtainable not be adequate for a clear and certain conclusion, let the ignorance be frankly recognised and not obscured by a cloud of conjecture, for it is a very great advance to know wliat it is we do not know. Tentative theories undoubtedly have their uses, but they have their abuses, for they are particularly liable to be confused with facts and to bias an otherwise impartial judgment and observation. They may lead to a wrong line of investigation. If one cannot refrain from evolving a theory and publishing it before it is fully proved (and it may often be desirable to do so), at least let it be clearly labelled as a tenta- tive theory and as nothing more, so that it will not be used for further deductions, even by its originator. With such caution the physiographer in Queensland has a wide field of useful work before him and probably a rich harvest of valuable scientific information. Without that caution he will only make confu- sion worse confounded and hinder the advance of real knowledge. PRESIDENTIAL ADDRESS. 17 LITERATURE REFERRED TO. Andrews, E. C.— Preliminary Notes on the Geology of the Queensland Coast, with reference to the Geography of the Queensland and New South Wales Plateau. Proc. Linn. Soc., N.S.W., 1902. Andrews, E. C. — Geographical Unity of Eastern Australia in Late and Post-Tertiary Time, with Applications to Biological Problems. Journ. Royal Soc., N.S.W., 1910. Andrews, E. C. — Plains and Peneplains of Australia. Commonwealth Year Book, 1912. Ball, L. C. — Certain Iron Ore, Manganese Ore, and Limestone Deposits in the Central and Southern Districts. Geological Survey of Queensland, Publication 194. Ball, L. C. — Notes on a Short Tour in the Gladstone District. Qld. Government Mining Journal, May, 1916. Danes, J. Y. — On some Problems of Queensland Hydrography. Queensland Geo- graphical Journal, 1909-1910. Danes, J. V. — A Tour along the Dividing Range from Aramac to Pentland. Qld. Geographical Journal, 1909-1910. Danes, J. Y. — On the Physiography of North-Eastern Australia. Proc. Royal Bohemian Society of Sciences, 1911. David, T. W. E. — Notes on Some of the Chief Tectonic Lines of Australia. Journ. Roy. Soc., N.S.W., 1911. ' David, T. W. E. — Geology of the Commonwealth. Federal Handbook of Australia, 1914. Davis, W. M. — Preliminary Report of a Shaler Memorial Study of Coral Reefs. “Nature,” 15th April, 1915. Davis, W. M. — Peneplains and the Geographical Cycle. Bulletin Geological Society of America, Sept., 1922. Dunstan, B. — Notes on the Geological History of Keppel Bay. Geological Survey of Qld., Publication 190. ' Hedley. C., and Taylor, T. G. — Coral Reefs of the Great Barrier. Aust. Assoc. Adv. Science, 1907. Hedley, C. — A Study of Marginal Drainage. Proc. Linn. Soc., N.S.W., 1911. Hedley, C., and Richards, H. C. — Report to Great Barrier Reef Committee. Qld. Geographical Journal, 1923. Jardine, F. — The Physiography of the Lower Fitzroy Basin. Qld. Geographical Journal, 1923. Jensen, H. I. — Geology of the Glass House Mountains and District. Proc. Linn. Soc., N.S.W., 1903. R.S. — C. 18 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Jensen, H. I. — Geology of the Volcanic Area of East Moreton and Wide Bay Districts, Queensland. Proc. Linn Soc., N.S.W., 1906. Jensen, H. I. — Some Geological Features of Northern Australia. Proc. Boy. Soe.,. Qld., 1922. Marks, E. 0. — Notes on Portion of the Burdekin Valley. Proc. Boy. Soc., Qld., 1912. Poole, W. — Physiography of North Queensland. Aust. Assn. Adv. Science, 1909. Beid, J. H. — Coal Measures of Western Moreton District. Qld. Govt. Mining Journal, Dec., 1922. Bichards, H. C. — The Volcanic Bocks of South-Eastern Queensland. Proc. Boy. Soc., Qld., 1916. Bichards, H. C. — Problems of the Great Barrier Beef. Qld. Geographical Journal, 1920-22. Taylor, T. G. — The Physiography of Eastern Australia-. Commonwealth Meteoro- logical Bureau, Bulletin No. 8. Wearne, B. A., and Woolnough, W. G. — Notes on the Physiography of West Moreton? Queensland. Journ. Boy. Soc., N.S.W., 1911. Vol. XXXVI., No. 2. 19 A Suggestion for a Biological Laboratory on Strad- broke Island for the Protection of Ceratodus. By Thos. L. Bancroft, M.B. (. Read before the Royal Society of Queensland , 28th April , 192%.) There are no young Ceratodus in the Burnett River at the present time. There are many aged fish, some of which spawn every year. The last week in August and first in September is the spawning season. Some fish continue dropping a few eggs right into November. Fish, particularly the Jew fish, eat up the Ceratodus spawn, every bit they can find ; yet owing to the cunningness of Ceratodus in placing some of the spawn at the extreme edge of the bank amongst water weeds and grass roots, in places inaccessible to other fishes, a fair quantity of it escapes. Unfortunately, however, as soon as the little Ceratodus hatch out they are devoured immediately by dragon-fly larvae. Under artificial conditions in a hatchery the little fish thrive well for several weeks ; they then cease eating and become rapidly emaciated ; air bubbles form in the stomach, and most of them die. I have likened this condition to teething in the human infant. A very lew, only, survive this illness. They commence feeding again and their heads rapidly grow, so that at about three months the head is three times the size it was. They now resemble in appearance tadpoles ; they have passed the critical period, and provided you can protect them from their enemies, dragon-fly larvae especially, they will continue to thrive. I am of opinion that from in-breeding the constitu- tion of Ceratodus has suffered. The Burnett is a comparatively small river; it is possible that no Ceratodus in it alive to-day has not fre- quently been, during flood time especially, up and down the river ; and that brothers and sisters have bred together for millions of years. One can imagine what the constitution of the human race would be under similar conditions. Fortunately Ceratodus is not confined to the Burnett only ; it exists also in the Mary River. I have longed to put this experiment to the test : — Mate some female Ceratodus from the Burnett with males from the Mary River, and vice versa. I am confident that the progeny would have a robust constitution. The sexes can be distinguished by the size of the cloaca being large in the female and small in the male. 20 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. There are more male fish than female : three to one ; the largest fish are mostly females. I think the time has arrived when a supreme effort should be made to raise money to initiate a Ceratodus station on Stradbroke Island, for with the advent of the new settlers into the Burnett and Mary districts the fish is doomed to extermination in a few years. Protection of the fish in its native home is impracticable and impossible. My scheme for the initiation of a Biological Laboratory with the principal object of protecting Ceratodus and preventing its extinction has been submitted to the principal scientists in Australia, to the Univer- sities, and all the scientific societies, and has been endorsed by them all. I propose to use the Blue Lake, a large fresh-water lagoon about nine miles from Amity Point and seven and a-half from Dunwich ; introduce into it a hundred Ceratodus, some from the Burnett and some from the Mary Rivers. Plant water weeds round the edge of the lake in places, after digging away the bank to allow the water to come in, and to a depth of about 3 feet. The Ceratodus is principally vegetarian in diet, and there are practically no water weeds at present in the Blue Lake. Instal a petrol engine to pump water from the lake to a hatchery and a number of large ponds, wherein to study the habits of the fish and rear young ones. I have prepared estimates of the cost of this undertaking, but have been asked to omit specific details from this paper, as the actual figures may depend on contingencies. I propose that control of the laboratory should be in the hands of the Professors of Biology of the Australian Universities. Vol. XXXVI., No. 3. 21 The Development of Buttresses in Queensland Trees. By W. D. Francis, Assistant Government Botanist. Plates I. -VI. Text-figures 1-7. ( Read before the Uoyal Society of Queensland, 28th April, 1924.) Contents. I. General remarks. II. A comparison of the distribution of buttressed trees in the Queensland and extra-Australian floras. III. The distribution of buttressed trees in the Queensland flora. IV. The stage of growth at which buttresses appear. V. The transition from roots to buttresses. VI. The number and dimensions of buttresses in various species. VII. The shapes of buttresses. VIII. The structure of buttresses. IX. Some characteristics of the roots and bark in buttressed trees. X. The extent to which root and stem respectively contribute to the construction of buttresses. XI. Conditions with which buttressed trees are associated. XII. The possible causes of buttresses. XIII. Summary. 1. — General Remarks. One of the most impressive features of luxuriant rain forests is the wide buttresses or plank buttresses which radiate from the bases of the stems of certain species of trees. These buttressed trees abound in dense rain forests in tropical and subtropical parts of the world. A. R. Wallace1 records them in the rain forests of the Amazon, T. F. Chipp2 in the West African rain forests, A. F. Schimper3 in the monsoon forests of Burma, H. N. WhitforcP in the rain forests of the Philippine Islands, J. Ii. Maiden5 in the rain forests of Northern New South Wales, and the writer6 in the rain forests of Southern Queensland. They are also known to occur in the forests of Java, Borneo, and New Guinea. From the foregoing statements it is evident that buttressed trees are found in all the continental land masses, which extend into the tropics, and in several of the large tropical islands. As the phenomenon is so widely distributed, it constitutes a problem of considerable bio- logical interest, and it is probable that a number of the observations and conclusions derived from an investigation of buttressed trees in Queens- land will be found to be applicable also to examples in other parts of the world. 22 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. II. A COMPARISION OF THE DISTRIBUTION OF BUTTRESSED TREES IN THE Queensland and Extra-Australian Floras. The occurrence of buttresses does not appear to have received general attention. In consequence, records of their distribution in Australia and abroad are very incomplete. The following table shows the distribution of buttressed trees in the Queensland flora and several extra-Australian floras so far as known to the writer. The extra-Australian portion of the table was compiled chiefly from the works of Schimper7 and IT. N. Ridley8, and from those of Chipp and Whitford, which are referred to in the preceding section. Queensland Buttressed Trees. Extra- Australian Buttressed Trees. Moraceae — Ficus' spp. Fig Trees. Moraceae — Ficus (Tropical Asia). Cecropia (Tropical America). Musanga (West Africa). Lauraceue — Endiandra discolor, Domatia Tree. Cunoniaceae — Weinmannia Benthami, Pink Marara. Weinmannia lachnocarpa, Marara. Ackama Muelleri, Rose-leaf Marara. Leguminosae — - Cynometra (West Africa). Piptadenia (West Africa). Parkia (West Africa). Meliaceae — Dysoxylum Pettigrewianum, Spur- wood, Satin-wood Dysoxylum sp. Red Bean. Cedrela toona var. australis, Red Cedar. Meliaceae — Dysoxylum (Java). Khaya (West Africa). Entandrophragma (West Africa). Anacardiaceae— Euroschinus falcatus. Ribbon -wood. Anacardiaceae — Dracontomelon (Philippine Ids.). El seocarpacese — Echinocarpus Woollsii, Carribin. Elceocarpus grandis, Quandong. Elceocarpus obovatus, Blue-berry Ash. Elceocarpus Kirtonii, Mountain Beech. Malvaceae — ■ Bombax (India, West Africa). Eriodendron (West Africa). Triplochiton (West Africa). Sterculiaceae — Tarrietia argyrodendron, Booyong, Hickory, Crow’s Foot Elm. Tarrietia actinophylla, Black Jack. Sterculiaceae — Sterculia (East Indies). Tarrietia simplicijolia (Malay Pen.) Tarrietia utilis (West Africa). BUTTRESSES IN QUEENSLAND TREES. 23 Queensland Buttressed Trees. Extra- Australian Buttressed Trees. Dipterocarpacese — Shorea (Malay Peninsula). Anisoptera (Malay Peninsula). Dipterocarpus (Philippine Ids.). Lophira (West Africa). Rhizophoracese — Ceriops Candolleana, a Mangrove. Rhizophoracese — Ceriops Candolleana (India). Anopyxis (West Africa). Combretacese — - Terminalia (West Africa). Myrtacese — Eugenia Francisii, Giant Watergum. Eugenia Luehmanni, Small -leaved Watergum. Syncarpia subargentea, Giant Iron- wood. Myrtacese — Eugenia (Philippine Ids.). ■ Verbenacese — Vitex (East Indies). Bignoniacese — Spathodea (Tropical Africa). The Natural Order Dipterocarpaceae, which comprises a large number of important trees widely and often densely distributed in rain forests of the Indo-Malayan region, is not represented in Australia, III. — The Distribution of Buttressed Trees in the Queensland Flora. The occurrence of buttresses in the more southern or sub-tropical rain forests of Queensland is more particularly under consideration, as the Northern rain forests of the State have not as yet been thoroughly investigated. The Queensland buttressed trees, with the exception of Ceriops Candolleana, which is a mangrove on the shores of the East Indies, tropical Asia and Africa, are rain-forest species. The majority of them belongs to genera which attain a relatively high numerical development of species in a region comprising India, tropical Eastern Asia, the East Indies, and New Guinea. Examples of these genera are : — Tarrietia, Echinocarpus, Elceocarpus, Dysoxylum, Euroschinus, Ficus, and Ceriops. All of the Queensland buttressed species observed by the writer with one exception belong to genera which are represented in the tropics to the North and North-West from Australia. The single exception to this qualification, Syncarpia subargentea, belongs to a genus which is repre- sented in tropical Queensland. 24 PROCEEDINGS OF THE ROYAL SOCIETY' OF QUEENSLAND. The table in Section II. shows that buttresses are not confined to any single Natural Order or circumscribed systematic group of plants, but that they occur in species and genera of Natural Orders often widely separated from one another. However, the very large systematic class of dicotyle- donous plants includes the buttressed species referred to in this paper. The same Natural Order may contain both buttressed and unbut- tressed species. In the Natural Order Sterculiaceas there are species of trees exhibiting singular contrasts in the shape of the bases of the stems. The two species of Tarrietia are very prominently buttressed, whilst the stem of the Queensland Bottle Tree ( Br achy chiton rupestre) is circular in transverse section and conspicuously bottle- or bulb-shaped. The Natural Order Malvaceae also contains examples of similar contrasts, but the trees providing them are chiefly found beyond Queensland. Belong- ing to this Natural Order, Bombax malabaricum is a buttressed tree in India and perhaps also in North Queensland, whilst bulb-shaped or tumid stems are known to be very strongly developed in the Baobab Trees ( Adansonia Gregorii of Western Australia and A. digitata of South Africa). Examples of unbuttressed, large, Queensland rain-forest species in Natural Orders containing buttressed species are the White Cedar ( Melia Azedarach) of Meliaceae, Callicoma serratifolia of Cunoni- aceie, and White Myrtle ( Rhodamnia argent ea) of Myrtaceae. IT. N. Whitford9 states that buttresses appear to be correlated with broadly developed crowns and with the dominance of certain trees over others. The occurrence of buttresses in Southern Queensland rain forests is not definitely associated with either of these conditions as a general rule. The following species are examples of trees which are sometimes dominant, often produce large spreading crowns, frequently attain large dimensions, and are generally unbuttressed: — Crow’s Ash {Flindersia australis), Cudgeree {Flindersia Bchottiana) , Yellow-wood ( Flindersia Oxleyana), Bogum ( Flindersia Bennettiana) , Marblewood ( Acacia Baheri), and Coondoo ( Sideroxylon Bichardi). A large number of species of the Natural Order Lauraceae, which is copiously represented in Southern Queensland rain forests, could also be cited as examples of the largest andi tallest trees which are without prominently developed buttresses. Some examples of the Lauraceae of this kind are : — Sassafras (Cinnamomum Oliveri ), Southern Maple ( Cryptocarya erythroxylon) , and Hard Bolly Gum {Beils chmiedia obtusif blia) . It should be remarked, however, that Southern Queensland rain forests differ from the Anisoptera-Strombosi formation of the Lamao Forest Reserve, where Whitford made his observations on buttresses. In the Southern rain forests of Queensland it is unusual to find repre- sentatives of a single Natural Order predominating in numbers over a large tract, as in the case of the Lamao Forest Reserve, in parts of which Whitford found that the Dipterocarpaceae predominated both in numbers and volume. Proc. Roy. Soc. Q’land, Vol. XXXVI. Plate I. Weinmannia lachnocarpa, rain forest. Cedar Creek, about 26.5° S. (Westward from Eumundi) . Diameter of stem and measurements of buttresses are shown in the Table in Section VI. Leaves of the palm, Archonto phoenix Cunninghamii, in background and a climbing aroid, Pothos Loureiri, on the tree stem in uppermost part of picture. An extension of a posterior buttress is shown on the left at about cr, e-fifth of height of picture. Photo.: W.D.F. [Face page 24.] Proc. Roy. Soc. Q;land, Yol. XXXYI. Plate II. Ficus eugenioides, rain forest, Kin Kin, about 26.2° S. (Eastward from G-ympie). The meshwork of aerial roots is evident in the picture. Photo.: W.D.F. [ Face page 24.] BUTTRESSES IN QUEENSLAND TREES. 25 Although it is convenient to describe certain species of trees as buttressed, or unbuttressed, the observer in the field is very frequently confronted with a large number of species exhibiting gradations between the buttressed and the unbuttressed conditions. In the more luxuriant rain forests the majority of the trees which are not positively buttressed lias flanged or somewhat perpendicularly flattened roots, Ceriops Candolleana is a small tree or shrub which flourishes on muddy shores of the sea, estuaries, and tidal streams. It has prominent buttresses which are well established in plants *9-1-2 m. (3-4 ft.) high. Of all Queensland buttressed trees the BoovOng ( Tarrietia argyro- dendron. var. trifoliolata) is one of the most common and most widely distributed, as it abounds in rain forests of both Northern and Southern parts of the State. As indicated in the comparative table in Section II., this variety belongs to a genus which is represented by buttressed species in the Malay Peninsula and West Africa. These facts concerning the distribution of the Booyong and its alliance with buttressed trees in distant parts of the world are of interest, as the buttresses of this variety are described in this paper in more detail than those of other trees. IV. — The Stage of Growth at which Buttresses Appear. The buttresses appear at a comparatively early stage of growth in a large number of rain-forest species. The dimensions of several examples of young buttressed trees are contained in the following table. Species. Diameter of Stem above Buttresses. Diameter of Stem. Maximum Diameter of Stem of Species. Height. Height. Maximum Height of Species. Tarrietia argyrodendron var. 7-2 cm. (2*86 in.) 1 9*9 m. (33 ft.) 1 trifoliolata 10 4 W einmannia lachnocarpa 6*3 cm. (2-5 in.) 1 8-4 m. (28 ft.) 1 TV 4*6 Echinocarpus Woollsii 8*8 cm. (3-5 in.) 1 10*3 m. (34 ft.) 1 sT 1 Endiandra discolor 5*6 cm. (2*25 in.) l 7-8 m. (26 ft.) 1 10 4*2 26 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. The measurements contained in this table indicate that buttresses are well established in comparatively young trees of these four species when they have attained from one-seventeenth to one-ninth of their maximum stem diameter and from one-quarter to two-ninths of their maximum height. These young trees were all very slender in habit and much below the height at which they develop spreading crowns or become exposed to the strain-producing action of wind. The young buttressed Tarrietia whose dimensions are shown in the above table had two buttresses, one measuring 30-2 cm. (12, in.) along the base and 7*6 cm. (3 in.) in perpendicular height, the other 22-8 cm. (9 in.) along the base and 8*2 cm. (3-25 in.) in perpendicular height. In discussing the development of buttresses in the Lamao Forest Reserve, Whitford has arrived at conclusions which are not apparent in Southern Queensland rain forests. He states1* : ‘ 4 It will be seen that the same causes which operate to produce short, thick boles in open places in temperate regions are possibly also present here, with the difference that the buttresses, which take the place and perform the function of uniformly thick trunks, appear later in the life of the tree. This means that the tree in youth, while it is crowded in the forest, develops a regular bole, but that when it reaches above the surrounding vegetation, in response to the increased heaviness of the top, the buttresses appear at the base, enabling the tree to withstand the extra strain. ’ ’ Y. — The Transition from Roots to Buttresses. The modification of roots in the process of forming buttresses has been observed by the writer chiefly in Tarrietia argyrodendron var. trifoliolata. In all instances it was noticed that the young buttresses consisted of the main surface roots which were flattened in a perpen- dicular plane where they joined the stem of the tree. At the distal end they gradually tapered into the rounded root. The young buttress could be compared to a triangle in which the base is long and the perpendicular height proportionately very short. With subsequent development the difference in the measurement of the base and the perpendicular height is lessened until the perpendicular height approaches or often exceeds the length of the base. Instances in which the perpendicular hight of the buttresses exceeds the length of their bases are very frequent in mature trees. This condition is commoner than the reverse in mature trees of Echinocarpus Woollsii ( see Plate V., fig, 2 and text-figure 4) and Endiandra discolor , for example. The following table shows the relationship of perpendicular height to length of base of buttresses at three stages of growth in Tarrietia argyrodendron var. trifoliolata : — Proc. Boy. Soc. Q-’land, Vol. XXXVI. Plate III [Face page 26.] Fig. 1. — Ficus macrophylla, rain forest, Imbii, about Big. 2. — Eugenia Francisii, rain forest, Kin Kin, 26.4° S. (Southward from Gympie). about 26.2° S. (Eastward from Gympie). Measure- ments of buttresses shown in Table in Section VI. Photos.: W.D.F. Proc. Roy. Soc. Q'land, Yol. XXXYL Plate IY. [Face page 26.] Fig. 1.— Tarrietia argyrodendron VAR. trifoliolata, Fig. 2.— Tarrietia actinophylla, rain forest, rain forest, Kin Kin, about 26.2° S. (Eastward from Macpherson Range, about 28.3° S. (Eastward from Gympie). Killarney). Photos.: W.D.F. f BUTTRESSES IN QUEENSLAND TREES. 27 Diameter of Stem of Tree above Buttresses. Perpendicular Heights of Buttresses. Length of Bases of Buttresses. Ratio of Average Perpendicular Height to Average Length of Base. 9-6 cm. (3-8 in) 8-2 cm. (3-25 in.) 10-1 cm. (4 in.) 8- 2 cm. (3-25 in.) 7-6 cm. (3 in.) 6-3 cm. (2-5 in.) 9- 5 cm. (3-75 in.) 20-3 cm. (8 in.) 22-8 cm. (9 in.) 20-3 cm. (8 in.) 17-8 cm. (7 in.) 17-8 cm. (7 in.) 22-8 cm. (9 in.) 41 : 100 21-0 cm. (8-3 in.) 45-7 cm. (18 in.) 48-2 cm. (19 in.) 30-5 cm. (12 in.) 15-2 cm. (6 in.) 122-0 cm. (48 in.) 76-2 cm. (30 in.) 50-8 cm. (20 in.) 60-9 cm. (24 in.) 45 : 100 60-5 cm. (2 ft.) (estimated) 320-0 cm. (10 ft. 6 in.) 289-5 cm. (9 ft. 6 in.) 289-5 cm. (9 ft. 6 in.) 396-2 cm. (13 ft.) 274-3 cm. (9 ft.) 243-8 cm. (8 ft.) 98 : 100 One of the young buttresses of the smallest tree whose dimensions are shown in the above table is represented by text-figures 2 and 3. The reverse process, in which the perpendicular height of the young developing buttresses greatly exceeds the length of their bases and in which subsequent growth by adding more to the length of the base tends to equalise the perpendicular and basal measurements, has not been observed by the writer. VI. — The Number and Dimensions of Buttresses in Various Species. The number of buttresses in any one species of tree frequently varies considerably. In a few cases only one or, less rarely, two buttresses is strongly developed. Instances in which three are well developed are much more frequent. Examples of small trees of Tarrietia argyroden- dron var. trifoliolata having respectively two and six relatively prominent buttresses are referred to in this paper. The mature trees of this variety exhibit a similar variation in the number of conspicuous but- tresses, and in some instances as many as eight buttresses may be developed. These observations are also applicable in a general way to Tarrietia actinophylla, Cedrela toona var. australis, Weinnmannia lachnocarpa, Eugenia Francisii, E. Leuhmanni, and Echinocarpus Woollsii. 28 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. The dimensions of the buttresses in four species of mature rain-forest trees are shown in the following table : — Species or Variety. Diameter of Stem above Buttresses. Perpendicular Heights of Buttresses. Length of Bases of Buttresses. Estimated Thickness of Buttresses. Tarrietia argyro- dendron var. trifoliolata 60-5 cm. (2 ft.) (estimated) 320 cm. (10 ft. 6 in.) 289 cm. (9 ft. 6 in.) 289 cm. (9 ft. 6 in.) 396 cm. (13 ft.) 274 cm. (9 ft.) 244 cm. (8 ft.) 6-3 cm. (2 ft. 5 in.) Echinocarpus Woollsii 60-5 cm. (2 ft.) 427 cm. (14 ft.) 408 cm. (13 ft. 6 in.) 229 cm. (7 ft. 6 in.) 229 cm. (7 ft. 6 in.) 183 cm. (6 ft.) 183 cm. (6 ft.) 152 cm. (5 ft.) 142 cm. (4 ft. 8 in.) 5-7 cm. (2-25 in.> Eugenia Francisii 106-6 cm. (3 ft. 6 in.) (estimated) 437 cm. (14 ft. 4 in.) 437 cm. (14 ft. 4 in.) 345 cm. (11 ft. 4 in.) 345 cm. (11 ft. 4 in.) 437 cm. (14 ft. 4 in.) 437 cm. (14 ft. 4 in.) 366 cm. (12 ft.) 274 cm. (9 ft.) 426 cm. (14 ft.) 335 cm. (11 ft.) 274 cm. (9 ft.) 183 cm. (6 ft.) 10-1 cm. (4 in.) Weinmannia lachnocarpa 120-9 cm. (4 ft.) 514 cm. (17 ft.) 302 cm. (10 ft.) 514 cm. (17 ft.) 196 cm. (6 ft. 6 in.) 635 cm. (21 ft.) 756 cm. (25 ft.) 514 cm. (17 ft.) 408 cm. (13 ft. 6 in.) 10-1 cm. (4 in.) VII. — The Shapes of Buttresses. In the paper by T. F. Chipp, which is referred to in Section I., the author discusses the outline of the outer edge (hypotenuse) of buttresses and the relative lengths of their base and height in West African trees, and applies these properties to the identification of species. The mea- surements given in Section V. show that the ratio of height to base in the buttresses of Tarrietia argryodendron var. trifoliolata is subject to variation as the tree develops. There are, however, more or less charac- teristic forms of buttresses which can be associated with several species of Queensland trees. The most conspicuous instances of this kind are Echinocarpus Woollsii and Endiandra discolor , which can be recognised very frequently by the edge (or hypotenuse) of the buttress curving outwards ( see text-figure 4 and Plate V., fig. 2). In Tarrietia argyro- dendron var. trifoliolata and T. actinophylla the edge of the buttresses is frequently almost linear or approximately straight ( see Plate IV.). On the other hand, the majority of buttressed Queensland species has edges which curve inwards towards the axis of the tree, as exemplified in Weinmannia lachnocarpa (Plate I.), Elceocarpus grandis (Plate VI., fig. 1), and Ficus spp. ( see Plates IT. and III., fig. 1). The relative thickness of buttresses is of some value in identifying species in a few instances. Comparatively thick buttresses are often found in Eugenia Francisii and comparatively thin ones in Echinocarpus Woollsii and Ficus spp. Prtoc. Roy. Soc. Q ’land, Vol. XXXVl. Plate \ . [Face page 28.] Fig. 1. — Cedrela toona var. australis, rain forest, Fig. 2. — Echinocarpus Woollsii, rain forest, Tam- Maepherson Range, about 28.3° S. (Eastward from bourine Mountain, about 28° S. (Southward from Bris- Killarney). bane). Photo.: TV.D.F. Photo.: J. E. Young. Pkoc. Roy. Soc. (J ’land, Yol. XXXVF Plate VI. [Face page 28.] Fig. 1. — Elaeocarpus grandis, rain forest, Traveston, Fig. 2. — Elaeocarpus obovatus, rain forest, about 26.3° S. (South-Eastward from Gympie). Macpherson Range, about 28.3° S. (Eastward from Killarney) . Photos.: W.D.F. BUTTRESSES IN QUEENSLAND TREES. 29 VIII. — The Structure of Buttresses. The depth of penetration of buttresses below the surface of the soil is comparatively slight. It is not uncommon to find parts of them with their basal extremities level with the surface or slightly above it, but more frequently they are embedded. In young buttressed trees with stem diameters of 7-2 and 9-6 cm. respectively, the buttresses penetrated the soil to a depth of about cm. (J in.) to 1-8 cm. (f in.). The buttresses in mature trees were observed to be embedded to a depth seldom exceeding 12-6 cm. (5 in.). These measurements can be expressed in terms of fractions by stating that the depth to which buttresses pene- trate the soil is approximately one-fifth of the perpendicular height of the buttresses in young trees and approximately one-twelfth of the per- pendicular height in mature ones. In mature trees with very large buttresses the fraction representing the proportion of depth of soil penetration to perpendicular height is probably about one-twenty-fifth. A vertical transverse section of a buttress from a specimen of Tarrietia argyrodendron var. trifoliolata measuring 50 cm. (20 in.) in stem diameter is shown in text-figure 1. This buttress had attained about one-third of the perpendicular height of the very large buttresses found in some specimens of this variety. The section was cut from the living tree in a perpendicular plane about 15-1 cm. (6 in.) from the stem. A vertical transverse section of a surface root of an unbuttressed species, Flindersia , Schottiana , is represented by text-figure 7. In contrast with the sections of buttresses its structure is almost concentric. The section of the larger buttress shows the extremely excentrie structure of the modified buttress-forming root, whose original centre is situated at a distance from the base of the buttress representing about one-twentieth part of the perpendicular height of the buttress where it was sectioned. The small buttresses of young trees also exhibit a pro- nounced excentrie structure as shown in text-figure 3. In the young Tarrietia with a stem diameter of 7-2 cm., the dimensions of whose two buttresses are given in Section IV., the original centres of the but- tressed roots were situated at a distance from the base of the buttresses equal to one-twelfth and one-twenty-fourth respectively of the perpen- dicular heights of the buttresses. This extremely excentrie structure of the buttresses in this small tree clearly indicates that the unilateral development in the roots of the alternating bands of tissue systems, which produces the buttress phenomenon, originated at a stage of growth very much earlier than that at which the tree was measured. The buttresses of the young trees of the other three species whose dimensions are shown in the table in Section IV. also exhibited a highly excentrie structure in transverse section. The curving lines appearing in the sections repre- sent some of the more conspicuous bands of wood parenchyma. They are not intended to be interpreted as “ annual rings/’ The lines inter- secting them indicate some of the more prominent wood rays. Gr. Haberlandt11 has figured a transverse section of a buttressed root of Parkia africana. 30 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Text-figures 1-6. 1, Vertical transverse section of buttress of Tarrietia argyrodendron var. trifoliolata X 2, Buttress of young tree of foregoing variety which had attained 9-6 cm. in stem diameter X i- 3, Vertical transverse section of same buttress X 4, Diagram of basal part of stem of Echwtocarpus Woollsii X 5, Transverse section of stem of same tree at the soil surface X 6, Transverse section of stem of same tree immediately above buttresses X J-g. S, surface of soil. BUTTRESSES IN QUEENSLAND TREES. 31 Some vertical sections of buttresses of Ceriops Candolleana were also made. They were similar to that of the young Tarrietia shown in text- figure 3, thereby indicating that the structural development of the buttresses is probably not very different in each ease. In all instances of prominently buttressed mature trees, which have been examined by the writer, a decided diminution was observed in the size of the stem or axis of the trees from just above the buttresses towards its point of contact with the soil. The extent of this diminution was measured in a large stump of E chino carpus Woollsii ( see text-figures Text-figure 7. Transverse section of surface root of Flindersia Schottiana X S, surface of soil. 5 and 6). The stem above the buttresses measured 604 cm. (2 ft.) in diameter, whilst its maximum diameter at the surface of the soil as the woody axis of the buttresses was only 22-6 cm. (9 in.), or three-eighths of its diameter above the buttresses. A similar tapering of the woody axis towards the surface of the soil was observed in large buttressed trees of Cedrela toona var australis , Tarrietia argyrodendron var . trifolio- lata, and Weinmannia lachnocarpa It was also evident in a specimen of Ceriops Candolleana. This peculiar feature of the structure of buttressed trees is the reverse of the form generally found in the stems of trees, which as a rule gradually taper from the ground upwards. Several examples of comparatively young buttressed trees of Tarrietia argyrdy dendron var. trifoliolata were examined, and it was ascertained that the characteristic diminution in the size of the stem towards the earth was not evident in trees up to 20-16 cm. (8 in.) in stem diameter above the buttresses. A buttressed specimen of this variety attaining the fore- going dimensions was sawn off at the surface of the ground and it was found that its stem or woody axis had not undergone appreciable diminution towards the base. Some microscopical examinations of the secondary wood in buttresses of Tarrietia argyrodendron var. trifoliolata indicated that this product contained tissues similar to those composing the stem in this variety. 32 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. IX. — Some Characteristics of the Roots and Bark. Large buttressed trees Avhieh have been uprooted by violent means, such as storms, generally show an extensive development of surface roots and no strongly developed tap-root, but the young Tarrietia with six buttresses and a stem diameter of 9*6 cm., which is referred to in Section V., had a large tap-root measuring 6-3 cm. (2-5 in.) in diameter. Schimper1- states that, owing to the prejudicial effect of humidity on the formation of cork, bark is only slightly developed on most of the tree stems in rain forest. The present writer has often observed that the bark is much thinner on the buttresses of rain-forest trees than that on the unbuttressed parts of their barrels. This diminution in the thick- ness of the cortex is often most marked towards or at the outermost edges of the buttresses. The measured thickness of the cortex on the stems and buttresses respective!}7 of several species of trees is shown in the following table : — Species. Diameter of Stern. Thickness of Cortex of Stem. Thickness of Cortex of Buttress. Thickness of Buttress Cortex. Thickness of Stem Cortex. Tarrietia argyrodendron var triloliolata 50-4 cm. (1 ft. 8 in.) 1-2 cm. (-5 in.) •48 cm. (-10 in.) 1 2-5 Echinocarpus Woollsii . . 60-48 cm. (2 ft.) 1-6 cm. (-625 in.) •63 cm. (-25 in.) 1 2-5 Weinmanma lachnocarra 70-56 cm. (2 U . 4 in.) •78 cm. (-312 in.) •48 cm. (TO in.) 1 1-6 X. — The Extent to which Root and Stem Respectively Contribute to the Construction of Buttresses. From the reversed or downwards taper of the lower part of the stem, vdiich is characteristic of strongly buttressed, large trees, the respective contributions of the root and stem can be deduced to a certain degree. This reversed taper or downwards diminution of the stem can be readily and naturally explained by assuming that the deficient portion of the louver part of the stem has passed into the buttresses. Hence that stage of growth in buttressed trees which exhibits the beginning of the downwards diminution of the stem or axis indicates the period at which the stem has commenced to participate in the construction of the but- tresses. As no appreciable diminution towards the base was observed in the stems of Tarrietia argyrodendron var. trifoliolata measuring from 15-12 cm. (6 in.) to 20-16 cm. (8 in.), it is concluded that the roots alone contributed to the construction of the buttresses in the stages of growth represented by these instances. The pronounced downwmrds diminution of the central stem of mature trees of this variety from above the but- tresses towards its insertion in the soil, as stated in Section VIII., shows that the axis loses a large proportion of its normal dimensions through BUTTRESSES TN QUEENSLAND TREES. 33 contributing to the development of the buttresses. It is, therefore, con- cluded that the stem performs a very large part in building up the buttresses in the latter part of the life of the tree. Corresponding to the foregoing order of modification of the root and stem in the development of buttresses is the extension of the sides of the buttresses collateral with the root and stem respectively. This relationship is evident in the table in Section V., which shows that in the young trees of Tarrietia argyrodendron var. trifoliolata the ba&e greatly exceeds the perpendicular height, whilst in mature trees the perpendicular height almost equals the length of the base. The table also indicates that the perpendicular height of the buttresses only shows a slight proportional increase upon the length of the base in stages of growth up to 21 cm. (8-3 in.) in stem diameter, whilst in the subsequent stage from 21 to 60-5 cm. in stem diameter a high proportional increase in the perpendicular measurement takes place. These facts conform with the respective shares attributed to the roots and stem in construct- ing buttresses, as they prove that the positive development of the buttresses proceeds along the sides respectively collateral with the root or stem during the stage of growth when the one or the other is computed to be the more active contributor. The extent to which these conclusions are applicable to examples other than those mentioned in this section is a subject requiring further investigation. XI. — Conditions with which Buttressed Trees are Associated. It has been remarked by Schimper13 that buttresses are a pecularity of trees in a tropical climate with abundant rainfall, and that the amount of rainfall necessary for their appearance is not yet ascertained. The present writer14 has shown that it can be definitely stated that in Queens- land the phenomenon appears in rain- forest trees when the average annual rainfall approximates, equals, or exceeds 60 inches, and that it is not confined to tropical forests but occurs in Queensland in relatively temperate climates at an altitude of 3,500 in latitude 28-2 degrees south. The occurrence of buttresses further south in Eastern Australia has been recorded by J. H. Maiden15 who has observed them in the Dorrigo Forest Reserve in about latitude 30-3 degrees. It is to be expected that in tropical rain forests in which buttressed trees abound the amount of the prevalent rainfall would be greater than that occurring in Southern Queensland rain forests on account of the more rapid evaporation in the tropics. High relative humidity and deep shade are conspicuous accompani- ments of rain forests containing buttressed trees. Unfortunately the humidity in Queensland rain forests does not appear to have been deter- mined. A high relative humidity is also a condition of the environment of the buttressed mangrove Ceriops Candolleana, which grows on tropical and subtropical sea coasts. R.S. — D. 34 PROCEEDINGS OF THE ROYAL SOCIETY QF QUEENSLAND. An extensive development of surface roots is also a noticeable feature of rain forests in which buttressed trees abound. These surface roots often arise from unbuttressed as well as buttressed trees. They are more conspicuous in the. heavy rain forests with a relatively high annual rain- fall as at Kin Kin (57 in.) than in those where the vegetation is less luxuriant and the rainfall much lower, as at Goodna (37 in.). XII.— The Possible Causes of Buttresses. Schimper16 remarks that the physiological causes of buttresses and their significance to the life of the tree are still obscure. In discussing the excentric development of plant organs, of which buttresses are examples, Haberlandt17 states that it is probable that the distribution of mechanical strains within an organ plays a leading part in producing asymmetrical thickening. A similar conclusion is assumed by Whitford, as the quotation from his work in Section IV. of this paper expresses the view that the buttresses appear at the base of the tree in response to the increased heaviness of the top, thus enabling the tree to withstand the extra strain. Although buttresses are not specially mentioned, Pfeifer,18 in discus- sing the stimulus of tension, remarks that the pull of the aerial parts upon the roots must act as a stimulus strengthening the latter, and that in general the mechanical demand largely determines the degree of development of the organs of attachment. The hypothesis which attributes the origin of buttresses to the direct effects of strain or tension induced by the upper, aerial parts of the tree appears to be the most obvious explanation of the phenomenon, but there are circumstances which are not in accordance with this view. As shown in Section IV., buttresses appear in many young trees long before they have attained the height at which they develop large crowns or become exposed to the strain-producing force of winds. Of all conditions with which buttressed trees are associated, high relative humidity appears to be the most characteristic, and it may possibly be a principal factor in causing the phenomenon. Other con- ditions which accompany buttressed trees, such as heavy rainfall, heavy crowns, and strain-producing factors such as wind, are also present in some temperate regions but do not induce the phenomenon there. But- tresses are essentially characteristic of the tropics and subtropics. In conjunction with a heavy rainfall in these regions, the solar heat, which directly increases the vapour pressure of water and the moisture-carrying capacity of the air, is known to profoundly stimulate vegetation, which luxuriates in the humid atmosphere and abounds in denseness and diversity of form unequalled elsewhere. In Section XI. the association of buttressed trees with rain forests in which surface roots are prominent has been remarked upon. It is not improbable that surface roots anteceded buttresses, which are essentially perpendicularly extended forms of surface roots. Besides humidity, BUTTRESSES IN QUEENSLAND TREES. 35 conditions which probably tend to develop surface roots are the moisture of the surface layer of rain-forest soils, the absence of much direct sun- light on the soil surface, and the fertile, humus-containing, upper layers of rain-forest soil. In addition to the nutriment it produces, the rela- tively high humus content of the soil towards the surface, because of its avidity for oxygen, would retard the oxygenation of the subsoil and exercise a further tendency towards the concentration of roots in the superficial layer of the soil. It would appear that in those cases where buttresses have been evolved the upper part of the principal surface roots has acquired an aerial character and is subject to some of the conditions of growth operating in stems. The perpendicular elongation of stems is a very prominent characteristic of the trees and erect shrubs of rain forests in which buttresses abound, and it is attributed to the attractive agency of light (phototropisin), acting in conjunction with the normal upward growth in opposition to gravity (negative geotropism). The upper part of the principal surface roots in buttressed species may be affected by negative geotropism and phototropisin either directly or indirectly through the stem ; and in this manner the perpendicular extension, which constitutes buttresses, may arise. Roots are most commonly subterranean organs, but in rain forests the roots of many plants possess an aerial or sub-aerial character. The adaptation of roots to an aerial environment is facilitated by the high relative humidity of the air and the exclusion of a great amount of direct sunlight in rain forests. These two conditions, therefore, are probably factors of considerable importance in the production of buttresses. The prevalence of the buttress phenomenon in trees whose roots have assumed a definite aerial character is exemplified in the epiphytical species of Fig trees (Ficus), which are so common in Queensland rain forests. In all the large specimens of the various species of these trees which the writer has seen the flattening of the roots in a perpendicular plane near the surface of the soil was prominent. It is also possible that hereditary factors are active, as several species of buttressed trees, such as the Figs (Ficus spp.), tend to retain their buttressed character when planted in parks and gardens away from their natural environment of the rain forest. In these cases, how- ever, the buttresses are not so large and conspicuous as in trees of corresponding sizes in the rain forests. The causes of buttresses, whatever they are assumed to be, are evidently to a great extent selective in their action as their effects are seen in species often wddely separated from each other in the classified systems in which plants are arranged, whilst species allied to those in which buttresses are strongly developed remain unbuttressed, although subjected to the same external conditions. 36 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. XIII.— Summary. Buttresses occur in Queensland in certain species of trees belonging to that section of the flora which is closely allied to the flora of the Indo- Malayan region. They are not confined to any single Natural Order of plants, but occur in species and genera of Natural Orders which are often widely separated from each other. Buttresses are well established in young trees of Weinmannia lachno- carpa, Tarrietia argyrodendron var. trifoliolata, Endiandra discolor, and Eckinocarpus Woollsii when they have attained from one-seventeenth to one-ninth of their maximum stem diameter and one-quarter of their maximum height. In young trees of Tarrietia argyrodendron var. trifoliolata the length of the base of the buttresses greatly exceeds their perpendicular height. In subsequent development the difference in the measurements of the base and perpendicular height is lessened until the perpendicular height approaches, equals, or exceeds the length of the base in mature trees. The depth to which buttresses penetrate the soil varies from about one-fifth of the perpendicular height of the buttresses in young trees to about one-twenty-fifth of the perpendicular height of the buttresses in mature trees with very large buttresses. Vertical sections of buttresses of Tarrietia argyrodendron var. tri- foliolata show that their structure is extremely excentric. * In a buttress, which had attained one-third of the maximum height of buttresses in this variety, the original centre of the root was situated at a distance from the base of the buttress represented by about one-twentieth of the perpendicular height of the buttress. In all instances of prominently buttressed mature trees, which were examined, a decided diminution was observed in the size of the stem or axis of the tree from above the buttresses towards its point of contact with the soil. In a specimen of Eckinocarpus W oollsii this reversed taper of the stem or axis of the tree reduced the axis at the soil surface to three- eighths of its diameter above the buttresses. It is computed that in Tarrietia argyrodendron var. trifoliolata the roots alone contributed to the structure of the buttresses in trees up to 20-16 cm. (8 in.) in stem diameter, and that in subsequent development the stem performs a large part in building up the buttresses. Conditions with which buttressed rain-forest trees are associated are a heavy rainfall, high relative humidity, the absence of much direct sunlight on the soil surface, and an extensive development of surface roots. In discussing the causes of buttresses the following factors are suggested as possible originating or contributing agents: — High relative humidity, the absence of much direct sunlight on the soil surface, BUTTRESSES IN QUEENSLAND TREES. 37 the humus content of the upper layer of rain-forest soil, negative geo- tropisin and phototropism. The possible activity of hereditary factors is also suggested. The causes of buttresses are evidently to a great extent selective in their action. AcKNO WLEDG MENTS. The writer is indebted to Mr. C. T. White, Government Botanist, for providing facilities for obtaining some of the photographs with which this paper is illustrated ; to Mr. IT. A. Longman, Director of the Queens- land Museum, and Professor E. J. Goddard, of the Queensland University, for the loan of literature ; to Mr. L. S. Twine, of the Queens- land Forest Service, and his staff, for assistance in obtaining some of the measurements which are quoted; and to Mr. J. E. Young for the photograph in Plate V., figure 2. The photographic illustrations originally appeared in “The Queenslander,” and the writer is indebted to the Editor of that Journal, Mr. W. J. Buzacott, for the process blocks which are used in this paper. REFERENCES. 1. A. R. Wallace, “Travels on the Amazon/’ p. 17, 1911. (Ward, Lock and Co., publishers. ) 2. T. F. Chipp, Kew Bulletin, No. 9, p. 268, 1922. 3. A. F. W. Schimper, “Plant Geography,” trans. W. R. Fisher, revised and edited, Groom and Balfour, tig. 140, p. 301, 1903. 4. H. N. Whitford, “Philippine Journal of Science,” ’Vol. 1, No. 4, p. 419, 1906. 5. J. H. Maiden, “Agricultural Journal of N.S.W.,” V., pp. 599-633, 1894. 6. W. D. Francis, Proc. Roy. Soc. Q ’land, Yol. XXXIV., p. 213, 1923. 7. Schimper, loc. cit., p. 305. 8. H. N. Ridley, “Flora of Malay Peninsula,” Vol. 1, 1922. 9. FI. N. Whitford, loc. cit., pp. 419-420. 10. H. N. Whitford, loc. cit., p. 420. 11. G. Haberlandt, “Physiological Plant Anatomy,’’ trans. M. Drummond, p. 677, 1914. 12. Schimper, loc. cit., p. 305. 13. Schimper, loc. cit., p. 304. 14. W. D. Francis, loc. cit. 15. J. H. Maiden, loc. cit. 16. Schimper, loc. cit., p. 305. 17. Haberlandt, loc. cit., p. 676. 18. W. Pfeifer, “The Physiology of Plants,’1 Ewart’s Translation, Vol. II., p. 128, 1903. 38 Vol. XXXVI., No. 4. On the Synonymy of Australian Flies belonging to the Genus Actina (Stratiomyiidae). By G. H. Hardy. (Read before the Royal Society of Queensland , 28th April , 1924.) In my catalogue of the Australian Stratiomyiidae (1920) there were listed the names and the synonymy of species of this genus as then understood, but I did not feel satisfied concerning the manner in which the names had been applied to the various recognised species. The material from the mainland of Australia then being studied contained mostly old specimens and the condition of these was inferior, so as to make difficult the determination of their specific values. In more recent investigations fresh specimens from Victoria, New South Wales, and Queensland were used, and an outstanding structural character was found to be persistently different in Tasmanian and main- land forms. This character consists of the proportional length between the first two joints of the antennae, the difference being greater in those from Tasmania. Such a character may be advantageously used to separate the island from the mainland species and, although the limits between two distinct Tasmanian species have been defined, it does not seem practicable to achieve similar results in the case of possibly distinct species from Australia, as all the forms examined grade too intimately. With the exception of A. cost at a White, a species that is remarkably persistent in colouration, the forms vary widely, as has been mentioned by White. The thorax may be metallic green or blue, a mixture of these colours ranging’1 to copper, or even practically black. The scutellum is similarly metallic and its colour may invade the otherwise yellow spines to any extent from the base towards the tip, but these spines are never entirely metallic as in A. costata. No form of the female has been found with an entirely blackish abdomen, which character is quite common amongst males and is a character of both sexes in A. costata. The males vary towards the typical colouration of the females and the females towards that of the males, so that where these variations of the sexes overlap it is not easy to determine the sex at a glance. The sexes are typically dimorphic in colour, as well as in structure, and invariably the legs of the female are yellowish, while those of the male are more or less stained with black. In an extreme case of variation the male has a dark border around the otherwise light dorsal area of the abdomen and transverse dark lines across this lighter ground. The female is typically light, and specimens from Tasmania are without or sometimes with only a very narrow dark border, but invariably have similar dark transverse bands. The Australian females are similarly marked, but the dark border is usually present and generally much broader. Such specimens AUSTRALIAN FLIES OF GENUS ACTINA. 39 taken in copula, as I have so far seen, also convince me that there is hut one species each from Australia and Tasmania amongst these varia- tions. Males of the same kind conjugate with females or various forms and vice versa. In considering the localities of the types of the various proposed species it will be found that Macquart described incisuralis, female, from Australia in 1847, and the male from Tasmania in 1850; evidently confusing two forms. Further, he described two males, filipalpis and fusciventris, and a female nitidithorax from Tasmania in 1850. The sex of the Tasmanian female may be erroneously marked, as Macquart refers to the colour of the legs as being dark, or perhaps the species is not an Actina as now considered, but belongs to some other genus. White (1914) described two species from Tasmania, incisuralis and costata, whilst Hill (1919), finding the Australian form to be distinct from the Tasmanian incisuralis of White, described his specimens under the name victorice, accepting White’s identification as correct. I have examined a paratype male of A. victorice in Dr. Ferguson’s collection and find this to be a form approaching the male variation, already referred to, in which the yellow area is present on the abdomen, but, however, restricted to a triangle on each of three segments. Such a variation is insufficient to warrant specific recognition. Enderlein (1921), like most other European authors on Australian Diptera, has overlooked the publications of recent Australian entomolo- gists, and he has described filipalpis and nigricornis from Australia and Tasmania respectively, apparently again dealing with the same two species. The first of these is evidently erroneously identified, because Macquart ’s species is from Tasmania, whilst Enderlein ’s is from Aus- tralia, and the second adds a further synonym to the commonest of the two Tasmanian species. The above remarks explain the synonymy given below, and to this a key is added for the determination of species as now understood. In judging the length of the antennal joints, care must be taken, as there are individual variations in these lengths. Specimens can be found from both Australia and Tasmania in which the first joint is as long in one as in the other, but in such cases the second joint is proportionally graded so as to bring about the same relative proportions as indicated in the key. KEY TO THE SPECIES OE ACTINA. 1. First joint of the antennae about (or less than) one and a-half times the length of the second. Sexes normally dimophic with regard to colour. Australian . . . . . . incisuralis Macq. First joint of the antennae at' least twice the length of the second. Tasmanian . . . . . . .... 2 2. Scutellar spines entirely metallic blue or green. Both sexes of a dark colour. Mountain form . . . . . . . . costata White. Scutellar spines yellowish at least at the apex. Sexes normally dimorphic with regard to colour. Lowland form filipalpis Macq. 40 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. SYNONYMY OF SPECIES. Actina incisuralis Macquart. Beris incisuralis Macquart, Dipt. Exot. suppl. 2, p. 28, 1847 (nec Macquart 1850). Walker, List Dipt. B.M. v. suppl. 1, p. 12, 1854. Actina incisuralis White, Proc. Lin. Soc. N.S. Wales, xli. p. 77, 1916 (part). Actina victories Hill, Proc. Lin. Soc. N.S. Wales, xliv. p. 450, figs. 1919. Hardy, Proc. Roy. Soc. Tasmania, p. 41, 1920. Actina filipalpis Enderlein, Mitt. Zool. Mus. Berlin, x. p. 191, 1921; (nec Macquart 1850). Hab. — Queensland, New South Wales, Victoria, South Australia, and Western Australia. Actina filipalpis Macquart. Beris filipalpis Macquart, Dipt. Exot. suppl. 4, p. 41, PI. iii., fig. 2, 1850 (nec Ender- lein 1921). Beris incisuralis Macquart, ibidem, p. 42, 1850 (nec Macquart 1847). Actina incisuralis White, Proc. Roy. Soc. Tasmania, p. 50, 1914; and Proc. Linn. Soc. N.S. Wales, xli. p. 77, 1916 (part). Hardy, Proc. Roy. Soc. Tasmania, p. 41, 1920. Beris fusciventris Macquart, ibidem, p. 42, 1850. Beris nitidithorax Macquart, ibidem, p. 41, PI. iii. fig. 3, 1850. Actina nigricornis Enderlein, Mitt. Zool. Mus. Berlin, x. p. 191, 1921. Hab. — Tasmania. Actina costata White. Actina costata White, Proc. Roy. Soc. Tasmania, q). 51, 1914; and Proc. Lin. Soc. N.S. Wales, xli., p. 77, 1916. Hardy, Proc. Roy. Soc. Tasmania, p. 41, 1920. Hab. — Tasmania. Vol. XXXVI., No. 5. 41 On a New Species of Melaleuca (Family Myrtaceae) from Southern Queensland. By E. Cheel and C. T. White. (One Text-figure.) (. Read before the Royal Society of Queensland, 28th April, 1924.) MELALEUCA GROVEANA sp. nov. Arbor, ramulis juvenilibus angular ibus; foliis alternis anguste lineari — lanceolatis mucronulatis crassis planis pellucido-punctatis glabris trinervis, nervis lateralibus margini per approximate, nervis secundariis obliquis obscuris ; spicis primum terminalibus, rhacide glabro ; calycis tubo glabro, lobis suborbicularibus ; petalis juvenilibus purpura- scentibus tandem albidis calycis lobis duplo longioribus ; filamentis phalangibus basi breviter connatis ; capsulis suborbicularibus vert-ice leviter contractis. A tree with rather close, flaky, slightly compressed fibrous bark of a light brownish colour, the very young shoots silky hairy, otherwise glabrous in all parts. Branches slightly angular when young but becoming more or less terete with age. Leaves linear-lanceolate 2|-4J cm. (1-lf in.) long, 3-6 mm. (1^-3 lines) broad, slightly mucronate, tri-nerved, the two lateral nerves forming an intramarginal vein about 1 mm. (-J line) dis- tance from the margin, secondary nerves rather oblique, not always discernible, oil glands plainly visible on both sides in dried specimens. Flowering spike terminating the branchlets, about 2 cm. (f in.) long, with 8-14 flowers, the rhachis and bracts when very young finely pubescent but all parts soon glabrous. Bracts 3-5 mm. (2.-|-2-| in.) long, with a prominent central nerve. Calyx-tube glabrous about 2J-3 mm. (about 2 i lines long) 1J mm. (nearly 1 line) diameter, crowned with 5 brownish sub-orbicular lobes about 1 \ mm (nearly 1 line) in diameter. Petals purplish when in bud; probably becoming creamy- white slightly tinged with purple when older but not seen in a fresh state, about twice the length of the calyx lobes. Stamens about 4 mm. (2 lines) long, creamy white, the filaments arranged in five bundles on short claws, the claw not exceeding 2 mm. long. Style from 3 to 4 mm. long (1J-2 lines) long, stigma simple. Ovary pubescent, 3-celled. Fruits orbicular or nearly so, with a slightly contracted orifice, 7-8 mm. (34-4 lines) diameter when fully matured, the valves rather deeply sunken. 42 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Text-figure. Melaleuca Groveana sp. nov. A, Shoot, natural size. B, Blower. C, Fruit, top figure in transverse section. NEW SPECIES OF MELALEUCA (FAMILY: MYRTACEiE). 43 Dab. — “Top of a high dry volcanic ridge near Edenvale Railway Station, Nanango District,” Queensland, collected by C. IT. Grove (No. 132), September, 1919. In systematic position the new species comes closest to M. Deanei F. Muell., and the fruits are almost identical with some specimens at present roughly grouped under that species, but as typical specimens of Mi Deanei have a woolly rbachis and the calyx tube is also more or less woolly it seems wise to keep them distinct ; the rhachis and calyx lobes, except when very young, appear always to be cpiite glabrous in the pro- posed new species. It has also close affinities with M. leiocarpa F. Muell., but this has smaller leaves and fruits. In general appearance the herbarium specimens very closely resemble those of Callistemon rugu- losus DC, flavo-virens Cheel., but the filaments are distinctly united at the base and grouped into five distinct bundles, so that it must be regarded as distinct from that plant. Specimens of timber show the sap-wood to be of a pale pink colour and the heart-wood to be very dark dull red, close-grained and hard. It, would probably be very durable. 44 Vol. XXXVI., No. 6. The Geology of the Silverwood-Lucky Valley Area. By Professor H. C. Richards, D.Sc. and W. H. Bryan, M.Sc., Department of Geology, University of Queensland. With a Geological Map, two Geological Sections, one Block Diagram, and Plates VII. -XX. {Read before the Royal Society of Queensland , 26th May, 1924 ). (I.) Introduction and Advance Summary Page. . . 44 (II.) Previous Work and Literature . . 46 (III.) Physiography . . 52 (IV.) The Silverwood Series (Devonian) . . 55 (V.) The Fault Block Series (Permo-Carboniferous) . . 60 (VI.) Correlation of the Fault Block Series . . 68 (VII.) The Walloon Series (Jurassic) . . 71 (VIII.) Igneous Activity . . 78 (IX.) Earth Movements and Trend Lines . . 77 (X.) Historical Summary . . 81 (XI.) Petrology . . 84 (XII.) Palaeontology . . 97 (XIII.) Economic Geology . . . . 103 (XIV.) Conclusion . . 105 I. INTRODUCTION AND ADVANCE SUMMARY. The following account is an attempt to elucidate and describe the geological features of a somewhat complex area in the south-eastern portion of Queensland and almost on the political boundary between that State and New South Wales. The area studied covers about 120 square miles of the parishes of Rosenthal and Wildash, and is 170 miles distant by rail from Brisbane. The nearest towns are Warwick (some 10 miles to the north) and Stanthorpe (28 miles to the south), and that portion of the main Southern Railway line which connects these towns intersects the area to be described, and, incidentally, provides a number of very interesting and instructive sections in its cuttings. There are three railway stations in or near the area. These are Morgan Park at the northernmost extremity, Silverwood in the central west, and Cherry Gully at the extreme south. Of these, Silverwood is the most convenient railway centre for those wishing to visit the* area. THE GEOLOGY OF THE SILVER WOOD-LUCKY VALLEY AREA, 45 46 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. The most useful maps of the area are the Warwick Land Agent map (40 chains = 1 inch) and the Glengallan Shire map (60 chains = 1 inch). The latter, however, does not cover all of the area under consideration. In order that the significance of the geological work carried out by earlier investigators in this area may be appreciated, a short advance summary of the conclusions arrived at by the present authors is necessary. (See Geological Map and Section A.) The oldest rocks of the district are a series of andesitic lavas and tuffs which include in their uppermost part a fossiliferous horizon made up of lenses of limestone. Above the andesites is a great thickness of banded radiolarian cherts and cherty shales. These two natural groups together form the Silverwood Series of the authors. This is of Devonian age and has been closely folded and heavily faulted. The faulting has. preserved a number of blocks of fossiliferous shales and grits, associated with which are acid tuffs and flows, the whole being of Permo-Carboni- ferous age. Lying unconformably on these two Palaeozoic series one finds shales and sandstones which are the local representatives of the very extensive coal-bearing “Walloon Series’’ of Jurassic age. In the southern part of the area the Silverwood Series is intruded and highly altered by the Stanthorpe Granite of Permian age. II. PREVIOUS WORK AND LITERATURE. The first event of geological interest recorded from the area was the discovery of gold, which took place, according to W. B. Clarke1 at Lord John Swamp in the year 1851 (although the official record as it appears in the Queensland Mineral Index2 refers the discovery to the following year). This eminent pioneer was the first geologist to visit the Silverwood-Lucky Valley area. In his Tenth Report he distinguishes between “the auriferous schists,” “the fossiliferous limestone,” “the carboniferous formation,” “the trappean alluvia,” and “the granite.” He also noted the structural and geological similarities of the Lord John Swamp, Canal Creek, and Pike’s Creek areas. The principal object of Clarke’s survey seems to have been the investigation of the possibility of the Darling Downs proving a productive goldfield. As a result of his study of the region he came to the conclusion that gold would not be -found in large quantities, but that the greatest mineral asset of the area was what he terms ‘ 1 The Coal Field of the Condamine. ’ ’ In the year in which Clarke published this report (1853), S. Stutchbury,3 working north from the Liverpool Plains district, passed over what is now the border of New South Wales and Queensland, followed the course of Canal Creek, and worked thence to Emu Creek; but although he passed quite close -to the Silverwood-Lucky Valley area he does not appear to have actually entered it. Further discoveries of gold were made in the neighbourhood in the years which followed Clarke’s visit, and in 1869 an area of 25 square miles was proclaimed and gazetted as the Lucky Valley Goldfield. In the same year the goldfield was examined and reported upon by Mr. THE GEOLOGY OF THE SILVERWOOD-LUCKY VALLEY AREA. 47 C. D’Oyly H. Aplin,4 the first Government Geologist for South Queens- land. This worker noted the resemblance of the rocks here met with to those of the Canal Creek, Thane’s Creek, and Talgai Goldfields, Avhich lie some 18 miles to the west, west-north-west, and north-west respectively. The interest attached to his interpretation of the geology of the district, together with the inaccessibility of the paper to most geologists, warrants a somewhat lengthy quotation from Aplin ’s report. “Silurian Rocks — Fossiliferous Beds. “The silurian rocks of this neighbourhood consist of shales, slates, sandstones, and calcareous grits. They have a general strike of N. 30 W. with a westerly dip of sixty or sixty-five degrees. It is probable that they belong to the upper silurian series, from the appearance of some fossils I was fortunate enough to meet with on the Avest bank of Elbow Gully about a-quarter of a mile from the river. “One set of beds (blue slates weathered white on the surface) were completely honeycombed with the cavities, from which branched corals and smooth crinoidal stems had dis- appeared. Above these beds were calcareous grits, filled exclusively with shells and casts of various BracMopoda, Gas- teropoda, and bivalves. Amongst them species of Productus and Spirifer were the most prominent; but with the most careful search I could find no trace of trilobites or graptolites. In general aspect they resemble the fossils of the diorite slates of Gympie. These beds form a narrow band betAveen the green- stone area and the river, and can be traced (with their contained fossils) for 2 miles along the northerly extension of their strike. ‘ ‘ Shales. “West of the diggings the slates and sandstones are much altered by intrusions of trap and greenstone, as Avell as by the proximity to the granite, and appear as hard banded crystalline schists near Lord John Swamp and on the range of hills on the Avest side of it. To the north-Avest, at a distance of 3 or 3^ miles from the granite, these altered slates give place to brown and dove-coloured rubbly shales. Beyond this the gold-bearing strata Avill probably not be found to extenci. . . . 1 1 Limestone. “Several outcrops of a greyish-Avhite crystalline limestone occur outside of, but near to, the S.W. boundary of the green- stone. They do not appear to have any definite or uniform direction, as would be expected if they were interstratified with the slates. They contain magnesia and are invariably associated with a rock of serpentinous character Avhich covers the Avhole area in which they are situated but is nowhere exposed in section. 48 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. “Presence of Intrusive Rocks — Greenstone and Hornblende Trap at Lucky Yalley. “Surrounding the Lucky Yalley diggings is an area of about 2 square miles, which is for the most part occupied by greenstone and hornblende trap which have been intruded amongst the slates to an extent which makes it impossible to define their several limits with any approach to accuracy.” In addition to these rocks Aplin’s map shows the presence of “Carboniferous sandstones and conglomerates” in the north-eastern portion of the area and the Stanthorpe granite in the south. Aplin’s silurian rocks included both the Silverwood Series (Devonian) and the infaulted Permo-Carboniferous series, all the fossils mentioned being found in the latter. This inability to recognise two distinct Palaeozoic series in the area existed until a few years ago, and although different workers have given different names to the palaeozoic rocks of the area they have been regarded as forming one series only and are still shown as such in the last official map of the Geological Survey of Queensland. The “Carboniferous sandstones” in Aplin’s map are representatives of the Walloon coal-bearing series, which is now known to be of Jurassic age. In 1872, R. Daintree5 included all the older rocks “from the southern boundary of Queensland up to latitude 18 deg. south” in the Devonian. Further, he states that “At the Lucky Yalley diggings and at Apis Creek near the Apis Creek homestead peculiar casts of encrinital stems, probably belonging to the genus Actinocrinus, occur in a lydianized slate. The strike of this bed in both places is about north 30 deg. west, which line would connect the two outcrops, although nearly 300 miles apart.” A list of districts which Daintree gives “where Devonian rocks prevail” is headed by Lucky Yalley, followed by Talgai. In the year 1879, ten years after Aplin’s work, A. C. Gregory,6 after describing the “coast range” from Point Danger to the head of the Mary River as the principal development of Devonian rocks in south-eastern Queensland, states : “In addition to this coast range there are some isolated patches of Devonian rocks as at . . . Lucky Yalley and at the head of the Condamine River, the Talgai and Pikedale Goldfields. . . In discussing the fossils found in these Devonian beds, the same author writes: “At Lucky Yalley and the head of the Rosenthal Creek, in the Warwick district, casts of spirifer have been found in the Devonian slates and sandstones. . . .” A. C. Gregory thus considered all the palaeozoic rocks of the Lucky Yalley- Silverwood area as of Devonian age, the conclusion being based on fossils obtained from what we now know are infaulted blocks, of THE GEOLOGY OF THE SILVER WOOD-LUCKY VALLEY AREA. 49 Pernio-Carboniferous age. Further, he correlated these strata with those of the coast range which are now called the Brisbane Schists, and which are at present doubtfully referred to the Ordovician period. In 1892, Robert L. Jack,7 then Government Geologist for Queens- land, in reviewing the work of Aplin to which reference has been made, came to the conclusion that the finding of Productus and Spirifer, together with Aplin ’s statement that “In general aspect they resemble the fossils of the diorite slates at Gympie” does not support a Silurian age, for he writes : — “To this latter statement I should attach greater weight than to the determination of two genera of Brachiopoda, which range up to a much later date than Silurian. I could even believe, without difficulty, in the absolute identity of the Lucky Valley beds with those of the Gympie Formation, especially as similar slates containing corals and crinoid stems have proved to be quite characteristic of the Gympie Formation, as developed in the Rockhampton district. From this meagre evidence I regard the slates, etc., to the west and south of Warwick as belonging to the Gympie Formation rather than to the Devonian.’7 This deduction was remarkably sound, for the fossils and the strata in their immediate neighbourhood are undoubtedly of Permo-Carbonifer- ous age, but the assumption that all the other rocks of this area were of this age has not been justified. In the year 1892, S. B. J. Skertchly8, then Assistant Government Geologist, prepared a report “On the Geology of the Country round Stanthorpe and Warwick, with especial reference to the Tin and Gold Fields and the Silver Deposits. ’ ’ This observer followed Jack in referring “The great series of schists, slates, quartzites, sandstones, greywackes, and limestone to the Gympie System. ’ ’ The map which accompanied this report embraced the Silverwood-Lucky Valley area and shows approximate boundaries of the “Permo-Carboniferous (slates, limestone, etc.)” Series with the Stanthorpe granite on the south and the Walloon coal measures on the north and north-east, but there is no reference in the text to the area under consideration. The next reference to the area is contained in the report of E. C. Andrews,9 of the Geological Survey of New South Wales, on the “Drake Gold and Copper Field.” Andrews writes: — “In 1907 the following forms were obtained by Mr. J. E. Carne, Assistant Government Geologist, in the railway cuttings some 6 or 7 miles west of Warwick: — Productus lor achy tlicerus, G.B. Sby. Spirifera StoJcesi (?) Konig. Martiniopsis subradiata var Morrissi Eth. fil. Crinoid stem. R.s. — E. 50 PROCEEDINGS OF TPIE ROYAL SOCIETY OF QUEENSLAND. these are Pernio-Carboniferous, probably of Lower Marine age. On the Queensland Geological Map they are marked as Gympie, The rocks in’ which these types occur are indurated clays and shales interbedded with limestones, andesitic and felsitic tuffs and breccias and thin tlaking shales covered with worm burrows. The whole series is excessively folded and faulted and penetrated by the ‘tin granite.’ ” A plate containing a series of interesting sections, with the caption Sections in railway cuttings near Warwick, Queensland, shewing con- torted Lower Marine (?) rocks” appears in the same report. Some of these sections certainly represent part of the older ‘ ‘ Silverwood Series ’ ’ of the present authors, and this fact probably explains why Carne10 considered similar rocks in the neighbourhood of Tingha as Permo- Carboniferous, while L. A. Cotton11 regards them as being considerably older and comparable with the Ordovician slates of Berridale and Tallong. A. B. Walkom12 in 1913 published a series of palaeogeographic maps of New South Wales in Permo-Carboniferous times, in which he shows the Warwick area as sea during the Lower Marine stage, closely adjacent to the coast line in the Lower Freshwater and Upper Marine stages, and as a land area in the LTpper Freshwater stage. In 1917 Mr. J. ITarward, M.A., of Warwick, Queensland, who takes a keen interest in the geology of the surrounding districts, was much struck by a fine specimen of Eurydesma cordatum which had been brought into Warwick and presented to the Warwick Technical College. As a result of inquiries Mr. ITarward ascertained the locality from which the specimen had been brought, and, securing the finder of the original specimen as a guide, visited the spot and found not only the richly fossiliferous Eurydesma Beds but made the discovery of a bed containing very numerous fossil leaves of Glossopteris and Gangamopteris, specimens of which he presented to the Geological Survey of Queensland. (Both the Eurydesma Beds and the leaf beds occur in what the authors term the Stanthorpe Road Fault Block.) As a result of further excursions into the field Mr. Harward was rewarded by discovering fossils near the Condamine River and close to the main railway line (the Condamine Block and Eight-Mile Block respectively of the authors). In March, 1918, one of us (Richards), at the invitation of Mr. Harward, visited the area and collected fossils, and from that time onwards paid frequent visits to the area. Later in the same year, A. B. Walkom also collected fossils from the Stanthorpe Road Block. Mr. E. C. Saint-Smith,18 of the Geological Survey of Queensland, visited the area early in 1918 in order to inspect certain iron ore deposits at Elbow Valley which were subsequently prospected on behalf of the State Government. In the year 1920 Mr. L. C. Ball,14 Deputy Chief Government Geolo- gist of Queensland, visited the Lucky ATilley Goldfield “in connection with an attempt to resuscitate reef mining. ’ ’ THE GEOLOGY OF THE SILVER WOOD-LUCKY VALLEY AREA. 51 In an examination of a specimen of limestone-bearing rock from the shaft of the copper mine at Elbow Valley, Mr. Ball discovered a speci- men of Helioliies, which he kindly allowed the authors to examine. A section cnt from one rock specimen showed a fine example of Stromatoporella. Mr. Ball concluded from these fossils that “the massive limestones and the interbedded sedimentary and volcanic rocks are of Devonian age.” He stated further: “It is judged that we have here a small inlier of Devonian rocks, perhaps less than 2 miles across.” [This was thought to be the first discovery of Devonian fossils in the district, but the authors have, through the courtesy of Mr. H. A. Longman, Director of the Queensland Museum, found in one of the old records of this institution a list of fossils referred to the Devonian period (probably determined by the late Mr. Etheridge) and recorded as obtained from specimens of limestone from the Lucky Valley area.] In his report Mr. Ball also referred to the finding of Permo-Carbon- iferous fossils on the banks of the Condamine River (the Condamine Block of the present authors). In 1922 E. C. Saint-Smith15 again visited the district, in order to examine “the occurrence of white marble in the Elbow Valley area,” which he states to be of “probably Lower Marine (Permo-Carboniferous) age.” In the same year Dr. A. B. Walkom described and figured (Qld. Geol. Sur. Pub. No. 270) a number of fossil plants from the Lucky Valley area. At the Wellington meeting of the Australasian Association for the Advancement of Science, January, 1923, the authors presented a paper on the “Radiolarian Rocks in Southern Queensland;” also a short account of the several infaulted blocks of Permo-Carboniferous age, with provisional lists of their contained fossils. A paper by the authors on “Permo-Carboniferous Volcanic Activity in Southern Queensland” was read before the Royal Society of Queens- land in September, 1923, and is printed in the volume of the proceedings for that year. In March, 1924, E. C. Saint-Smith once more visited the area and furnished a report on “Certain Copper Deposits at Silverwood” (Qld. Govt. Min. Jour., 1924, p. 85). During the last few years, the authors have made many visits to the area, each of several days’ duration, with the object of elucidating this interesting and somewhat complex geological area, which it was considered might, on account of its geographical position, well serve as a key in elucidating the broader problems met with in the correlation of the late Palagozoic strata of Central and Northern Queensland with their better known equivalents in New South Wales. Silverwood- Lucky Valley Area. 52 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. THE GEOLOGY OF THE SILVER WOOD-LUCKY VALLEY AREA. 53 III. PHYSIOGRAPHY. Structurally the Silverwood-Lucky Valley area falls naturally into three distinct regions, each of which has its definite geological charac- teristics and its corresponding topographical forms. (See Block Diagram.) These units may be briefly summarised as follows: — 1. The north-eastern portion of the area is comparatively flat and rolling country, and made up of the almost horizontal Walloon Series of Jurassic age, the youngest beds of any consequence in the area. 2. The southern part consists of hilly country formed from the erosion of the Stanthorpe granite of Permian age. 3. The remaining portions of the area, in the centre and west, are occupied by the rugged ridges of the steeply dipping Palaeozoic sediments, the oldest rocks of the area. The. first of these structural units is a small section of the extreme south-eastern edge of the great stretch of rolling plateau country known as the Darling Downs. When examined from one of the convenient heights to the south or west, this unit appears to be a remarkably flat and homogeneous structure, and although as one descends, the apparently flat plains are seen to be made up of long rolling hills, the sense of unity remains. No abrupt hills or steep declivities, or sudden physio- graphical changes of any kind appear to break the almost monotonous uniformity. This is the least elevated region in the Silverwood-Lucky Valley area, and is drained by the Condamine River, which flows in meandrine fashion in a general north-westerly direction across the region under discussion. To all appearance the Condamine is quite independent of the nature and structure of the rocks over which it is flowing, but this independence may be more apparent than real. Davis10 has pointed out that “Regions of essentially horizontal structures normally have wandering streams, and the unit under discussion is such a region. At one point the Condamine leaves the easily corracled horizontal Walloon Measures and cuts across a projecting mass of hard Palaeozoic rocks. This small section of the valley presents quite a different appearance, being narrow and deep with high, steep, rocky slopes. The most probable explanation of this curious fact is that the stream has been superimposed on the hard Palaeozoic rocks as a result of the erosion and removal of an overlying veneer of Walloon sediments. The river has partly adjusted itself to the new conditions and flows for a considerable proportion of the gorge almost parallel with the strike of the Palaeozoics. It may be stated in passing that this gorge has been considered as a site for a dam to conserve the waters of the Upper Condamine. The Condamine is joined by a number of tributaries which drain the Palaeozoic country to the south and west. The more important of these are the Lucky Valley Creek and Elbow Valley Creek in the south- east, and the Lord John Swamp, Oaky and Eight-Mile Creeks, which. 54 PROCEEDINGS OF TPIE ROYAL SOCIETY OF QUEENSLAND. before reaching the Condamine, unite to form the Lord John Swamp. The general topography of this first unit suggests mature physiography in a homogeneous structure. The second structural unit is formed by the most northerly portion of the extensive “ Stanthorpe” granite, and consists of high hilly country which forms the source of many streams. The topography is typical of that of granite country, the hills being for the most part smoothly rounded and the valleys wide and open. In spite of the fact that it forms a rather important divide and provides the source of many streams, there is no evidence of deep dissection, and the topography characteristic of the headwaters of streams is here not developed. The Y-shaped valleys and other phenomena usually associated with head waters are to be found lower down the stream courses, where the contact-hardened Palaeozoic rocks are penetrated. A stranger to the area, if placed in the principal of these narrow valleys which is locally known as 1 ‘ The Gap ’ ’ and faced towards the “mature” topography of the granite, is only with difficulty persuaded that he is looking upstream. As is the case of the first unit, we are dealing with a mature topography in a homogeneous structure. The third physiographic unit is made up of the members of the Devonian “Silverwood Series” and the infaulted blocks of Permo- Carboniferous strata. This unit is part of a large area of similar structure stretching away for many miles to the west and north-west. It covers about half the area shown in the accompanying geological map, and is made up of hilly and rugged country. Structurally it is very different from the other units, both in its lack of homogeneity and in that the beds which make it up are for the most part standing almost vertically. As a result of these two factors the drainage in this unit is largely controlled by the underlying rocks. The streams are, for the most part, of the “subsequent” type, the general trend following the strike of the Silverwood Series (N.N.W.). This is very well illustrated by the Rosenthal Creek, Avhich, except for the Condamine, is the largest stream in the area. The long strike valley through which this stream flows is one of the most marked physiographic features of tile whole area, and as such is made use of by the Main Southern Railway, which climbs the valley almost to its head. The actual positions of the valleys in this region are controlled either by the fault planes of the great strike faults or are determined by the selective corrasion of the well- jointed radiolarian cherts in preference to the massive andesitic tuffs, which are consequently left as strike ridges. The topography of the region is that characteristic of the head waters of streams which have reached the mature stage of the geographic cycle, in that there is a remarkably close adjustment of streams to structures. Relationship of the Eucalypts to GeologiCxAl Formations. The relationship of the flora to the underlying geological formations was closely considered as far as the eucalypts were concerned, and much help was derived therefrom in mapping the area, especially in suggesting THE GEOLOGY OF THE SILVERWOOD-LUCKY VALLEY AREA. 55 where the junction lines were likely to go. The relative preponderance of certain species, also the presence or otherwise of others, was very marked. For example, the Stringy-bark ( Eucalyptus eugenioides ) occurred on the granitic rocks only, while the Gum Top Box ( Euca- lyptus liemiphloia) did not occur on the granite, but was found on both the Silverwood Series and the Walloon Series. The following table is of occurrences noted in the field, and the authors desire to thank the Government Botanist (Mr. C. T. White) for coming into the field and checking the determinations. Walloon Coal Silverwood Stanthorpe Measures. Series. Granite. Eucalyptus tereticornis (Blue Gum) X X. X Eucalyptus hemiphloia (Gumtop Box) X X — Eucalyptus maculata (Spotted Gum) X X X Eucalyptus crebra (Narrow leaf Iron Bark) . . X X X Eucalyptus paniculata (Iron-bark) X X X. Eucalyptus melliodora (Honey Gum) — X X Eucalyptus conica (Box) — - X X Eucalyptus eugenioides (Stringy-bark) if I X IY. THE SILVERWOOD SERIES (DEVONIAN). The oldest rocks of the area, though they present considerable lithological diversity, are, for reasons to be presented later, thought to represent continuous deposition confined to one period, and as such will be treated as one group, to which the authors have given the name “Silverwood Series.” These rocks cover, roughly, one-half of the Silverwood-Lucky V alley area, or about 60 square miles, but this is only one small portion of the total outcrop, for the series has been traced by one of us far to the west and north-west, and the total area of continuous or nearly continuous outcrops in South Queensland must be in the neighbourhood of 1,600 square miles. The total thickness of the Silverwood Series, after allowing for duplication by faulting, and assuming an average dip of 60 deg., is at least 11,000 feet. The series is closely folded and the dips are always very steep, so that 60 deg. may be regarded as quite a conservative estimate of the average. The strike, except in the immediate vicinity of the major faults, is remarkably constant, seldom varying much from N.N.W.- S.S.E. This direction is all-important in a consideration of the Silver- wood Series, for not only is it a reflection of the great axes of folding, but a number of great (strike) faults follow the same direction. Further, it has already been pointed out that as a close adjustment has been attained by the streams flowing over this series to the structure of the underlying rocks, the physiographic features of the present day emphasise this N.N.W.-S.S.E. trend. Lithologically the Silverw'ood Series may be roughly divided into two great groups. The lower of these is made up for the most part of 56 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. tuffs and sediments of an andesitic nature at least 6,000 feet in thickness. The upper portion consists of banded radiolarian cherts and fissile shales, the combined thickness of which is at least 5,000 feet.17 The sequence met with as one ascends the Siiverwood Series can now be dealt with in more detail. The lower portion of the Siiverwood Series is for the most part monotonously uniform, being made up of a succession of altered andesitic tuffs and lavas of massive character without cleavage and with very little sign of definite bedding. At two horizons only there occur breaks in the otherwise monotonous succession, and both these occur interbedded with the upper portions of the andesitic tuffs. Fortunately they form marked departures from the lithological type, and are thus of great value in the structural interpretation of the geological map. The lower of these is made up of discontinuous lenses of limestone, some of them of considerable size, which lie with their long axes parallel to the general direction of the strike. Lithologically they are in such decided contrast with the neighbouring andesitic tuffs that they form an easily recognis- able horizon for stratigraphical work of a local nature, but their value for purposes of more distant correlation is much increased by the fact that they are for the most part fossiliferous. ( See fig. 2, Plate IX.) Immediately above and closely associated with the limestone lenses is a rock which presents some very remarkable features. ( See fig. 1, Plate IX.) It is a curious agglomeratic rock, made up of rounded and sub-angular fragments of fossiliferous limestone identical with that of the underlying lenses, set in an andesitic tuffaceous groundmass. The rock is evidently of a very unusual type, but a similar rock has been described by Benson18 from the Tamworth area, where it is closely associated with the “Nemingha” limestone. Further, the authors have specimens of an almost identical rock from Moongan, Struck Oil, near Mount Morgan, but very little is known of the geology of the immediate locality. Almost at the top of the andesitic tuffs and some 600 feet above the limestone lenses is another distinct horizon. This is formed by a coarse conglomerate, and differs radically from all the other members of the Siiverwood Series. It is only some few feet in width, but sporadic outcrops have been found over an area of 30 square miles, and these display great uniformity, showing that the conglomerate is not merely a local development. The most marked features of the con- glomerate are, first, the nature of the beautifully rounded to semi-angular pebbles and boulders, ranging from about 2 inches to 8 inches in diameter, which are made up entirely of igneous rocks; second, the presence intermingled with these, of smaller and less regular fragments of limestone ; third, the setting of all these in a fine andesitic tuffaceous groundmass quite typical of the andesitic rocks in the lower part of the Siiverwood Series ; and fourth, the sudden, change and absence of any gradation between the coarse pebbles and the fine enclosing matrix. The igneous boulders include granitic, porphyritic, and andesitic rocks. THE GEOLOGY OF THE SILVERWOOD-LUCKY VALLEY AREA. 57 The source of these and the conditions under which this remarkable conglomerate was formed are matters of considerable interest, but neither problem has been solved to the satisfaction of the authors. The general nature of the deposit is very suggestive of glacial action, but no facetted or striated boulders have been observed, and in the absence of such evidence the authors are reluctant to accept this explanation. The lumps of limestone were probably portions of contemporary fringing reefs similar to those which earlier formed the limestone lenses referred to above. The upper portion of the Silverwood Series presents a marked contrast to the lower, both lithologically and structurally, for the massive character, lack of bedding, and absence of cleavage of the o*ie- is replaced in the other by beautiful banding and high fissility, while at the same time the tuffaceous sediments of the earlier deposits have been largely replaced by cherts and shales. {See Plate VII.) The junction between these two so dissimilar members of the Silverwood Series is marked by a breccia in which angular fragments of the (upper) banded rocks are embedded in the uppermost portion of the (lower) andesitic tuffs. The movements of the Silverwood Series will be dis- cussed in some detail in a later section, but one may mention here that there is much evidence for supposing this rock to be a crush breccia between the competent andesitic tuffs and the incompetent cherts and shales. Succeeding this breccia is a great series of cherts showing very pronounced banding and containing myriads of chalcedonic casts of radiolaria. {See Plate XI.) The banding, when examined under the microscope, is seen to result from the frequent and abrupt changes from a coarser lighter-coloured layer rich in tuffaceous material containing relatively few radiolarian casts, to a darker, finer, more cherty and non-tuffaceous layer in which the radiolarian casts are much more numerous. The width of the bands, both absolutely and relatively, one to the other, may vary considerably; but such variations, though marked, are not frequent, most of the numerous outcrops observed remaining fairly close to the average or type rock. {See Plate VIII.) These banded cherts are usually broken by numerous small faults, and this phenomenon becomes so marked in the immediate neighbourhood of the several great faults that the rocks sometimes present an almost brecciated appearance. Miniature step-faulting is especially common, and owing to the marked banding of the cherts is shown with diagrammatic distinctness. One result of the stresses to which the banded rocks have been subjected has been the fracturing of the more brittle cherty layers and the forcing of the tuffaceous material into the fractures, thus simulating some of the characteristics of igneous intrusions. {See Plate XIII.) As one ascends the series the banded cherts are gradually replaced by a rock which consists of fissile shales containing numerous interbedded layers of chert. The average thickness of these bands is about 2 inches, 58 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. and they form about one-third of the rock, the remaining two-thirds being occupied by the fissile shales. These rocks also give evidence of crumpling and contortion, but the evidence is less obvious. This probably is a reflection of the different nature of the rock and need not be interpreted as a result of a lessening intensity of stresses. It is shown chiefly by cross-fracturing, crumpling, and the frequent pinching out of the chert bands. This rock type is in its turn replaced by greenish-brown fissile shales containing no chert bands. (See fig. 2, Plate X.) Again, there is evidence of intense crumpling and contortion. No undoubted traces of radiolaria have yet been discovered in these shales, but an interesting feature, first noted by Andrews,19 should be mentioned. This is the development of worm tracks in such great numbers as to entirely cover the exposed bedding planes of certain strata. These vary in diameter between one-eighth of an inch and one inch. (See fig. 1, Plate X.) Age of the Silverwood Series. In spite of the crush-breccia at the junction, the andesitic or lower portion of the Series appears to be quite conformable with the overlying cherts, and as these grade into the fissile shales forming the highest- known members of the Series, there is good reason for regarding the whole as representing one continuous period of deposition. That the whole series is at least pre-Permo-Carboniferous in age is proved by the fact that a number of blocks of fossiliferous strata of this age have been faulted down into the Silverwood Series, and the further fact that pebbles of banded radiolarian chert identical with that of the Silverwood Series are found in the Upper Marine (Permo-Carboni- ferous) conglomerates at Eight-Mile Creek. (See fig. 3, Plate XI.) The only fossiliferous horizon so far discovered that could be used for determining the age of the Silverwood Series is that formed by the limestone lenses which have been described as occurring near the top of the andesitic tuffs. Lists of the fossils found and determined from the different outcrops in the area will be found in a later chapter. (See page 97.) It may be said of these lists that while they have, as a whole, a Devonian aspect, they differ from the better-known deposits of this age in Queensland and New South Wales in that they contain certain genera more characteristic of the Silurian period. Fortunately for our purpose the different limestone horizons of New England, N.S.W., have been studied in some detail, especially by Benson,20 and assigned to their relative stratigrapliical horizons, while lists of their contained fossils have also been published.21 The oldest and most persistent of these horizons in New England is the Nemingha limestone, which can be traced for 150 miles ; this is succeeded, after a considerable time interval, by the Loomberah lime- stone, which has a somewhat restricted development, while the highest horizon is occupied by the well-known Moore Creek limestone. Of the Loomberah limestone “very little is known of it as yet,”22 but of the other two, detailed information is available. THE GEOLOGY OF THE SILVERWOOD-LUCKY VALLEY AREA. 59 The evidence adduced by a study of this information ( see Chapter XII., “Palaeontology”) is distinctly in favour of a tentative correlation of the limestone horizon of the Silverwood Series with the Nemingha limestone of the Tamworth area. The Silverwood district would thus appear to have formed portion of the Eastern Province of Benson rather than the North-Eastern, in which all the Queensland developments have hitherto been placed. The correlation of the Silverwood with the Nemingha limestone is supported by other considerations, as will now appear. Attention has already been drawn to the remarkable agglomeratic rock, made up of fragments of fossiliferous limestone set in an andesitic tuffaceous groundmass, which immediately succeeds the limestone lenses of the Silverwood Series. An almost identical rock is found in the Devonian rocks of New England closely associated with the Nemingha limestone. It is found on no other horizon in either Series. The cherts which succeed the andesitic rocks of the Silverwood Series contain innumerable chalcedonic casts of radiojaria, but few of those examined under the microscope show any detailed structure. While many of them retain enough of their original outline to suggest that the majority of the radiolaria present belong to the sub-order Spumellaria , none have so far been generically identified. Hence no determination of the age of the contained beds can be made on these organisms. Litho- logically, however, these radiolarian rocks show a remarkably close resemblance to those of the Tamworth area, which have been made so well known by the work of David23, Pittman24, and Benson25. In each case the most conspicuous features are the numerous alternating bands of light-coloured tuffaceous material and darker very compact cherts. While the lithological resemblance of the Silverwood radiolarian rocks with those of Tamworth is so close as to justify a tentative correla- tion on this fact alone, were no better evidence available, it is important to remember that banded radiolarian rocks seem to be very similar in many different parts of the world and, moreover, they are found on many different horizons. In the present instance the lithological evidence of the cherts must be regarded as an important confirmation of the dossil evidence of the limestone, for the relationship existing between the limestones and the chert is much the same in Tamworth as in the Silverwood area. Benson places the Nemingha limestone in the lowest part of the Middle Devonian. Assuming that this is correct and that the Silverwood Limestone occupies the same horizon, the underlying andesitic portion of the Silverwood Series may be regarded as Lower Devonian and the over- lying Aadiolarian cherts as Middle Devonian, while the uppermost fissile shales may be Upper Devonian, thus corresponding with the mudstones r>f Tamworth and Barraba. 60 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Y. THE FAULT BLOCK SERIES (PERMO-CARBONIFEROUS), Faulted down into the steeply dipping strata of the Silverwood Series are blocks of marine and freshwater strata, with which is asso- ciated a great thickness of lavas and tuffs, for the most part of an acid nature. These blocks (see Geological Map) are four in number and have been termed by the authors : — I. Eight-Mile Block, in the north-western part of the area. II. Tunnel Block, in the south-western part of the area. III. Stanthorpe Road Block, in the central part of the area. IY. Condamine Block, in the eastern part of the area. These occurrences together form what may be termed the ‘'Fault Block Series ’ ■ of Permo-Carboniferous age. The fact that the Series is represented by a number of isolated blocks prevents the construction of a complete stratigraphical sequence. However, after determining the sequence in the individual blocks and comparing their faunas, a reasonable, if incomplete, sequence has been evolved. The individual beds which go to form the Series will now be considered in the order in which they are supposed to have been deposited. The Eurydesma Beds. The lowest representatives of the series which have been found are- in the central part of the Stanthorpe Road Block. This block forms an elongated outcrop about 2 miles in length and about 1 square mile in area. Immediately to the west of, and striking parallel with, the main Warwick-Stanthorpe road are found the lowest known members. They are separated from the representatives of the Silverwood Series lying to the east by a fault, so that they may not form the true base of the Fault Block Series. They are somewhat over 200 feet in thickness and are composed of conglomerates and grits containing numbers of marine fossils, the most typical genera of which are Fenestella, Dielasma, and small Pakeopectens. (For detailed lists of fossils of these and other beds the reader is referred to Chapter XII., page 100.) Succeeding these grits, etc., are brown and grey sandstones some 270 feet in thickness, for the most part poorly fossiliferous, but containing in the upper part C ardiomor pha gryphioides. There follows; a rather marked lithological change, the strata for the next 140 feet being for the most part darker in colour, liner in texture, and more calcareous. The first 20 feet are made up of fine dark-grey calcareous: sandstones, in part crowded with Chonetes. Following this bed is a similar rock some 50 feet thick, containing numerous fossils, mostly pelecypoda, the most abundant and typical of which is Eurydesma cord, ahum. This is regarded by the authors as one- of the key horizons of the series. Immediately over this all-important ,THE GEOLOGY OF THE SILVER WOOD-LUCKY VALLEY AREA. 61 bed one finds about 70 feet of dark, almost black, very compact, cherty shales, with occasional fossils, for the most part bryozoa. Above these is a second marked lithological change, the dark fine-grained rocks being followed by about 70 feet of breccias and grits containing fragments of large and thick-shelled pelecypods. Succeeding these is a bed of mud- stone about 20 feet in thickness, for the most part devoid of fossils. The above section (although complicated by a fault which causes some repetition of beds in the field) is continuous, and the members which compose it are known collectively to the authors by the term ‘ ‘ Eurydesma Beds, ’ ’ although Eurydesma itself has a quite restricted range in the field. The Eurydesma Beds occupy the greater part of the Stanthorpe Road Block, strike between N. and N.N.W., and for the most part dip steeply to the west, although in the immediate neighbourhood of the bounding faults the dip is often erratic. In the southern portion of the block and occupying considerably higher ground, on the flanks of what are known locally as the ‘‘Wallaby Rocks,” is a series of strata striking approximately N.E.-S.W. and dipping to the N.W. These will be referred to as the “Wallaby Beds.” The relationship which this development bears to the Eurydesma Beds is not clearly shown in the field. The marked divergence in strike may represent a time gap of some importance, but on the other hand it may represent a dislocation brought about by the heavy faulting which dropped the whole Stanthorpe Road Block into the Silverwood Series. The fact that the two sets of beds are adjacent units in the same infaulted blocks suggests that they are approximately of the same age. The palaeontological evidence on the one hand emphasises the break, for it shows that the lowest of the Wallaby Beds have been deposited under fresh water, but on the other hand the age of these beds as determined by fossil evidence shows them to be more clearly related to the Eurydesma Beds than to any other Permo- Carboniferous strata so far discovered in the whole area. Lithologically the Wallaby Beds as a whole differ markedly from the Eurydesma Beds, but the lowest of the former resemble rather closely the uppermost strata of the latter. In the light of this somewhat contradictory evidence the authors are of opinion that between the Eurydesma Beds and the Wallaby Beds there is probably a break in the sequence of Permo-Carboniferous strata in the area, but that this break is probably not of any great magnitude. Wallaby Beds. These beds measure about 400 feet in thickness. Although the individual members are perfectly conformable and although there is a distinct lithological uniformity, the beds fall naturally into two divisions of approximately equal thicknesses, the lower of which is characterised by freshwater fossils and the upper by marine fossils. 62 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. The basal beds are separated from the uppermost strata of the' Eurydesma Beds by a marked change in the direction of strike (amounting to nearly 60 deg.). These and the overlying strata for a vertical distance of 200 feet are made up of mudstones and shales, which when fresh are light or dark-grey in colour, but which weather readily into a reddish-brown colour by the oxidation of the contained iron. Fossil remains of Gangamopteris, Glossopteris, and Noeggerathiopsis occur throughout the whole 200 feet, but are only sporadically distri- buted in the lower portions. Towards the upper limit, however, they are very abundant. Here they occur as glossy films of iron oxide, in which all the details of the venation are clearly shown. (See Plate XIX.) (Some of these have been described and figured by Dr. Walkom).26 A count of recognisable fossils obtained from this horizon showed seventy-five specimens of Gangamopteris as against fifty-five of Glossopteris. Immediately above these freshwater beds and perfectly conformable with them is a horizon of grey shale; weathering into reds and browns, which is crowded with the glossy black casts of a pelecypod similar in general appearance to, but much smaller than, Astartila gemma. Above this horizon are developed nearly 200 feet of similar sandy shales and mudstones, all characterised by the reddish-brown colour of their outcrops, and for a great part crowded with fossil remains of marine organisms. Although lithologically this development is indivisible, it can be roughly divided into three sections according to the preponde- rating fossils. The lowest of these is characterised by a great develop- ment of productids, including Strophalosia. The middle division is notable particularly for the great abundance of Spirifers, while the uppermost portion is crowded with fossil Bryozoa. (See figs. 3, 4, 5, Plate XX.) The sequence of the Wallaby Beds is abruptly terminated by one of the bounding faults which have dropped the Stanthorpe Road Block down into the Silverwood Series, The continuation of the sequence thus broken must be based on palaeontological grounds, and here the evidence is not as definite as one would wish. However, in the opinion of the authors the nearest successors to the Wallaby Beds are developed in what they have termed the Eight-Mile Fault Block. This is the largest occurrence of Permo- Carboniferous material in the whole Silverwood-Lucky Valley district. It covers almost 8 square miles, and is found on the Eight-Mile Range and the creek of the same name about 8 miles south of the town of Warwick. Structurally the Eight-Mile Fault Block resembles the Stanthorpe Road Block in one important feature. It is divisible into two portions, the beds of which are nearly of the same age, but in which the strike of one section is markedly different from that of the other. This dividing line is almost certainly a fault, nearly meridional in direction, which separates the western half, which strikes N.N.W., from the THE GEOLOGY OF THE SILVER WOOD-LUCKY VALLEY AREA. 63 eastern half, in which the strike is W.N.W. The strata thus separated have been termed the ‘'Rhyolite Range Reds” and the “Bight-Mile Creek Beds” respectively. The strata most nearly related in age to the uppermost members of the Wallaby Rock Beds are to be found in the lowest members of the Eight-Mile Creek Beds. These provide a good, if discontinuous, series of outcrops along the creek which gives its name to the beds. The Eigfot-Mile Greek Beds. These beds strike about W.N.W. and dip 60 deg. to the N.N.E. Their total thickness cannot be determined with accuracy, but must be in the neighbourhood of 5,000 feet, although this estimate makes no allowance for possible but undetected strike faults. The beds, although perfectly conformable, fall naturally into two groups ; a lower, about 2,400 feet thick, made up of alternating shallow-water marine and freshwater deposits, and an upper, of approximately equal thickness, composed entirely of volcanic lavas and tuffs mostly of an acid nature. The basal members of the former group as exposed in this section are shallow-water grits and conglomerates, the former containing fragmentary remains of pelecypods and hrachiopods and the latter many large rounded boulders of banded greenish-blue cherts which have undoubtedly been derived from the radiolarian cherts of the Silverwood Series. For a considerable distance above this shallow-water marine horizon the outcrop is obscured by alluvial wash. When it is again found the section is made up of 1,200 feet of sandstones, black shales, and more coarse conglomerates containing boulders of banded radiolarian cherts. ( See fig. 3, Plate XI.) The black shales contain rather numerous remains of Glossopteris. (See fig. 1, Plate XX.) Although a search was made for specimens of Gangamopteris, not one was found. In view of the preponderance of Gangamopteris over Glossopteris in the Wallaby Beds less than 3 miles away, this is thought by the authors to have considerable stratigraphical significance. This succession is strongly suggestive of oscillations between shallow-water marine and freshwater conditions, and such an interpretation is supported by the fact that immediately following the last bed of black mudstone containing plant remains is a very uniform bed of sandstone 75 feet in thickness, containing numerous casts of Martiniopsis. (See fig. 4, Plate XIX.) This in turn is followed by another bed of black mudstone, which completes the section so far as normal sediments are concerned. Lying immediately above this group of fossiliferous beds is a great thickness of lavas and tuffs. These voleanics have already been dealt with in some detail by the authors, in a paper entitled “Permo- Carboniferous Volcanic Activity in Southern Queensland,”27 so that only a short description is needed here. The voleanics are made up of rhyolites, dacites, andesites, and basalts. ( See Plate XIV.) This order indicates not only the sequence of the eruptions but the relative THE GEOLOGY OF THE SILVERWOOD-LUCKY VALLEY AREA. 65 quantities of material ejected, the earlier rhyolites and dacites being greatly in excess of the later andesites and basalts. Interbedded with the rhyolite and dacite flows are great thicknesses of the corresponding tuffs. The total thickness of volcanics in this section, if one assumes an average dip of 60 deg., must be nearly 2,500 feet. This is not the only occurrence in the Silverwood-Lucky Valley area where one finds a great thickness of acid lavas and tuffs developed in the Permo-Carboniferous sequence. On the contrary, they form an important feature in at least three other occurrences. The thickness in each case is so great and the lithological character so uniform that one may reasonably assume that the different occurrences of the volcanics were the result of the same volcanic phase. The palaeonto- logical evidence of the accompanying sediments does not oppose such a conclusion. The Rhyolite Range Beds. In the same fault block as that in which the Eight-Mile Beds occur, but in the western portion, one finds the Rhyolite Range Beds forming a conspicuous physiographic feature and constituting part of the most important divide in the area. These consist of shallow-water sediments containing large paheopectens such as Aviculopecten mitchelli , together with examples of Martiniopsis and other marine fossils, overlain by a great succession of lavas and tuffs for the most part of an acid nature and lithologically identical with those previously described. {See Section B.) The Tunnel Fault Block. The Tunnel Fault Block covers almost 1 square mile, and forms an elongated outcrop stretching from the tunnel north of Cherry Gully railway station, to the Rosenthal Creek. This block also occupies high ground and undoubtedly ow^es its existence, like the Rhyolite Range, to the weather-resisting properties of the acid lavas constituting the greater part of its mass, at least 1,000 feet, which in this case overlie about 400 feet of coarse conglomerates and shallow-water grits containing occasional marine fossils. The Condamine Fault Block. Here, again, evidence of violent volcanic activity, mostly of an acid nature, is met with, but both the volcanic material itself and the associated sediments differ considerably from those other occurrences which have just been dealt with. In the first place the volcanic phase is represented by tuffs only, no lavas being present, and in the second place the sediments immediately underlying the volcanics contain no freshwater deposits and no typical shallow-water marine deposits. These apparent anomalies can be explained in terms of the greater distance of this particular occurrence from the shore line and from the volcanic centre, the nearest of the other developments of tuffs and lavas (the Eight-Mile Creek Beds) lying over 6 miles to the west. (The distance separating the two blocks may, of course, have been considerably greater before they were clown-faulted). R.S. F. 66 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. The Condamine Beds. The greater part of the Conclamine Fault Block thus probably represents a deeper water phase of part of the Permo-Carboniferous sequence, already dealt with. However, the lithological and palaeonto- logical differences are so marked as to warrant a moderately full description. The block is exposed for 1£ square miles., but probably extends further to the east, where it is overlain with a marked uncon- formity by horizontally disposed Walloon strata of Jurassic age. The strikes of the Condamine Beds vary between north and north-west, the dip being to the west and south-west, and ranging from 60 deg. in the lower beds to 40 deg. in the upper. Assuming 50 deg. as the average dip, a conservative estimate of the total thickness of Permo-Carboniferous material in this section would be 4,000 feet. Of this thickness 3,000 feet is composed of normal marine sediments, while the overlying 1,000 feet is made up of rhyolitic tuffs and very tuffaceous grits. The lowest members of the series visible in the field are dark-grey to black, fine, compact, somewhat calcareous shales. These are in part richly fossiliferous, one horizon, regarded by the authors as particularly important, being crowded with the remains of Trachypora wilkinsoni ( see figs. 1 and 2, Plate XVIII.). Numerous other fossils are present, including curious corals resembling Trachypora in habit and mode of growth, Zaphrentoids, and many erinoid stems. The total thickness of these fine dark fossiliferous shales must be approximately 2,800 feet.# Immediately above these is a very fine-grained massive black rock, almost basaltic in appearance, which micro-sections show to be really a tuffaceous quartzite. This is succeeded by a series of rocks of a distinctly different type. They are all lighter in colour and coarser in grain than the underlying beds, the former fact resulting from an admixture of rhyolitic ash with the normal sedimentary material, and the latter indicating a shoaling of the Permo-Carboniferous sea in this locality. Although there is some lithological variation, the whole group can be described as well- jointed rhyolitic tuffs with which are interbeddecl tuffaceous grits and conglomerates containing occasional marine fossils, the whole totalling almost 1,000 feet in thickness. An important fact is that the uppermost bed (a tuffaceous grit) contains marine fossils. Age of the Fault Block Series. The age of the series as a whole is easily determined. In an earlier paragraph it has been shown that Jack long ago suspected that part of the area was of Permo-Carboniferous age, and the occasional fossils found * In the Stratigraphical Table (Table 1) accompanying their paper on “ Radiolarian Rocks in Southern Queensland ” the authors place these fossiliferous shales above the “Rhyolitic Tuffs and Breccias.” This the authors now regard as unlikely. THE GEOLOGY OF THE SILVERWOOD-LUCKY VALLEY AREA. 67 since in various parts of the district have all gone to confirm his opinion. It was never recognised, however, that the fossils were obtained from isolated outcrops, hence the conclusion was reached that the whole area was occupied by rocks of Permo-Carboniferous age. This we have shown to be false, but the collections of the authors have confirmed beyond dispute that the faulted rocks themselves are of Permo-Carboniferous age. The authors’ interpretation of the probable stratigraphical relation- ship of the different members of the Fault Block Series is shown in the accompanying table (Table I.). This has been constructed partly from evidence in the field, but partly on the assumption that several key horizons occupy relatively the same position in the Silverwood-Lucky Valley area as they do in the Permo-Carboniferous Series of the Hunter River District of New South Wales. TABLE I. Probable Relationship of the Several Members of the Fault Block Series. Feet. Stanthorpe Road Block. Eight-Mile Block. Tunnel Block. Condamine Block. 6,000 5,000 4,000 3,000 2,000 1,000 f Marine ( Strophal ■ Wallaby | osia) Beds i Fresh- Water Gan- L gamopteris) f | . . . Eurydesma < horizon Eury- desma Beds j Marine l Volcanics Fresh- water Sediment s with Glossop- ieris inter- bedded with Shallow- water Marine Shallow - Water Marine Volcanics Volcanics Shallow- Wafer Marine. Volcanics interbedded with Shallow- Water Marin0 , . . . . Tra'hy- pora horizon Marine 68 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. VI. CORRELATION OP THE FAULT BLOCK SERIES (PERMO-CARBONIFEROUS). One of the main objects which the authors had in view in investi- gating the Silverwood-Lucky Valley area was the correlation of the late Palaeozoic rocks of Queensland with those of New South Wales. The systematic position of the different developments of strata of this age in Queensland, in their time relations, both to each other and to the corresponding formations in New South Wales, is and always has been eminently unsatisfactory. This is well illustrated by the variable significance attached to the term Gympie Series in the past and the wide differences of opinion as to the subdivision of the “ Bowen Series” at the present time.28 Since these late Palaeozoic formations have been worked out in considerably more detail and the different developments correlated with much more success in New South Wales than in Queens- land, the most reasonable method of attacking systematic classification of those deposits in the latter State is by reference to the stratigraphical sequence of the former. The value of such a method will undoubtedly be controlled by geographical conditions. The nearer the Queensland deposits to those of New South Wales the more direct and reliable should be the correlation between the two. In this connection the Silverwood- Lucky Valley area is almost ideally situated, being nearty on the Queensland-New South Wales border. Unfortunately, the nearest group of Permo-Carboniferous rocks in New South Wales (the Drake area) is the one about which the information is most scant and least definite. On the other hand, the Hunter River district, further to the south and some 300 miles from the Silverwood-Lucky Valley area, has been worked out in masterly fashion by Sir Edgeworth David,29 while details have been filled in by various later writers, with the result that the strati- graphical sequence in this area now forms an excellent basis for the present purpose. The following table (Table II.) indicates the conclusions at which the authors have arrived after comparing the incomplete record of the Silverwood-Lucky Valley area with the complete sequence as met with in the Hunter River district. Each section is on the same scale, but it is, of course, impossible to give exact values for the duration of the breaks in the local sequence Further, it should be noted that the Hunter River section is based on maximum thicknesses, while the Silverwood-Lucky Valley section is based on very conservative estimates. The correlation is based primarily on the assumption that Eurydesma cor datum and Trachypora ivilkinsoni occupy the same respective horizons in both areas, and that the predomi- nance of Gangamopteris over Glossopteris in the lower portion of the Wallaby Beds warrants a close comparison with the Greta coal measures. TEE GEOLOGY OF THE SILVER WOOD- LUCKY VALLEY AREA. 69 Feet 16,000 14,000 12,000 10,000 8,000 6,000 4,000 2,000 TABLE II. HUNTER RIVER, NEW SOUTH WALES. SILVERWOOD-LUCKY VALLEY, SOUTH EAST QUEENSLAND. Fert. 16, 000- UPPER COAL I Upper i Newcastle. Newcastle. Dempsey St age. MIDDLE | COAL 1 MEASURES UPPER MARINI-: 4 LOWER COAL MEASURES Tomago Coal. 14,000 12,000 Crinoidal 1 Shales J Volcanic: = of Iviama Eight-Mile Creek Beds. Muree Stage. — TRACEY PORA WILKINSONI VOLCANICS AND 10,000 UPPER FRESH WATER. 8,000 Bran 'ton Stage. Greta Coal. LOWER MARINE Farley Stage. Loeliinvar S' age. Condamine j Beds. I i UPPER y MARINE. Strophalona GANGAMOPTERIS predomi- nates over GLOSSOPTER1 S Wallaby 6,000 Beds. H LOWER ' FRESH WATER. 4,000 EUR\ DESMA CORDATE M Eurydesma Bed- LOWER MARINE. 2,000 70 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. That the scheme of correlation based on these three fundamental assumptions is upheld by the more general evidence provided by a consideration of the suites of fossils found in the different fault-blocks may be tested by a perusal of the lists of fossils given in a later portion of the work. The most important difference between the two sections, and one of great economic significance, is the absence of coal seams in the local record. The Greta coal appears to be represented only by richly fossiliferous freshwater shales, while the ecpiivalents of the Tomago and Newcastle coal measures have not yet been met with. Another interesting point of difference, though one of less economic importance, is the presenece of a great series of volcanic rocks in the Silverwood- Lucky Valley section equivalent in position to the Crinoidal Shales of the Hunter River sequence. This difference is readily reconciled, however, if we consider the Permo-Carboniferous section as displayed in the Kiama district of New South Wales, for here, too, the crinoidal shales are only poorly represented as such, their place being largely taken by the “Volcanic Series.” This correlation of the volanics of the Silverwood-Lucky Valley area with that of the Kiama district has already been dealt with by the authors in a previous paper, 30 and need not be elaborated here. Suffice it to say that while there is much evidence for considering the two series contemporaneous, they are chemically and petrologically quite distinct, the Kiama series being essentially potash-rich while the local volcanics are relatively poor in this base. [Note. — The authors are indebted to Mr. G. D. Osborne, B.Sc., for drawing their attention to the volcanic series of Murrurundi and Wingen, N.S.W., which Carne (“Kerosene Shale Deposits of New South Wales,” p. 259) regarded as equivalent in age to the Upper Freshwater Measures of the Permo-Carboniferous Series, on account of their close association with the Murrurundi kerosene shale. These volcanics are made up in ascending order of agglomerates, tuffs, rhyolites, basalt. The whole, according to Carne, are probably underlaid by the Greta coal measures.] With regard to the occurrence of rocks of Permo-Carboniferous age in the Drake area, the authors would like to offer a suggestion. Although the}^ have not had an opportunity of examining these rocks in the field, the authors have paid particular attention to the literature dealing with the area as containing the nearest Permo-Carboniferous deposits to the Silverwood-Lucky Valley district. All the sedimentary rocks of the area are grouped together under the term “Permo-Carboniferous,” but the descriptions of many of the unfossiliferous slaty rocks are applicable word for word to typical members of the Silverwood Series, which has been shown to be of Devonian age. The authors have related how the Silverwood-Lucky Valley area was long thought to be made up of Permo-Carboniferous rocks entirely, and have demonstrated how this fallacy arose from the collecting of fossils of this age from isolated fault blocks which had been let down into highly folded Devonian sediments. It seems possible that a similar misunderstanding exists with respect to THE GEOLOGY OP THE SILVERWOOD-LUCKY VALLEY AREA. 71 the neighbouring Drake area, and that the Permo-Carboniferous fossils have been collected from fault blocks in Devonian strata equivalent to the Silverwood Series. Carne31 certainly regarded some of the members of the Silverwood Series seen by him in railway cuttings to the south of Warwick as being Permo-Carboniferous in age. Further, he probably had these rocks in mind and those similar unfossiliferous strata in the Drake area when he expressed the opinion that the slates of the Tingha region were of Permo-Carboniferous age, although other workers regarded them as being considerably older.32 The suggestion that the Drake area, like the Silverwood-Lucky Valley area, may consist of blocks of Permo-Carboniferous strata down- faulted into Devonian sediments receives some support from the fact that Benson has described similar blocks of Permo-Carboniferous age faulted down into Devonian in the Nundle and neighbouring districts.33 Similarly, in the literature dealing with the various occurrences of Gympie ’ 1 rocks in Queensland it is. often difficult to reconcile the Permo-Carboniferous fossils and the descriptions of the fossiliferous beds with the rock descriptions of the area as a whole, and the present authors hold with considerable confidence the view that much of these so-called Gympie beds rightly belong to the Devonian or some older period. This confidence has been strengthened as the result of hurried visits by one or other of the authors, first to the area to the west and south-west of Warwick where radiolarian cherts associated with Devonian lime- stones were found at Texas, Gore, etc., and second, to the upper waters of Cressbrook Creek, between E.sk and Crow’s Nest, where a series of rocks were discovered identical with the lower portion of the Silverwood Series as displayed in the bed o'f Rosenthal Creek, in the type area. These regions are both marked as Permo-Carboniferous on the latest official maps. VII. THE WALLOON SERIES (JURASSIC). Between 35 and 40 square miles of the area under discussion are covered by a series of deposits which differ markedly in almost every respect from both the Silverwood and Fault Block Series. These beds occupy the north-eastern portion of the area and, consisting as they do of comparatively flat and rolling country present a marked topographical contrast to the remainder of the area, and form the first of the three distinct physiographic units which were considered in an earlier section of this work. {See Block Diagram.) Towards the centre of the area where they are associated with the older rocks already described, this series occurs only as isolated patches on the minor divides or as a thin discontinuous veneer which is rapidly disappearing as the result of erosion, but as one proceeds in a north-easterly direction they attain a very considerable thickness. Structurally they differ from the older rocks, since in all the outcrops observed they are horizontal or nearly so, and thus show very marked unconformities where they are found in contact with either the Silverwood or Fault Block Series. Although the 72 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND actual lines of junction cannot be seen, a comparison of the horizontal conglomerates and shales in the cutting immediately north of Silver wood railway station with the steeply dipping, compact, well-jointed shales of the Silverwood Series in the succeeding cutting leaves no doubt as to the unconformity which separates them, while the examination of the sections on the left bank of the Condamine, near the point where Elbow Valley Creek joins the river, shows that the unconformity with the Fault Block Series is quite as marked. Lithologically the series is made up of shales, sandstones, and conglomerates, all of freshwater types. The sediments have a fresh unaltered appearance, and this, together with the poor development of joints and absence of folding and faulting, gives to sections an appearance of newness which enables them to be readily distinguished from each of the other rock types met with in the area. The best natural section is displayed on the left bank of the Condamine River at the point already discribed, where the corrasion of the sandstones and shales has resulted in the production of a conspicuous bluff. Outcrops of the series may also be seen near Elbow Valley school,, on the low divides between Lord John Swamp Creek and Lucky Valley Creek and between Oaky Creek and Eight-Mile Creek, at the top of the divide on the Stantliorpe Road to the north of Cherry Gully railway station, and near Silverwood railway station. Of these outcrops the last is the most instructive, as the railway cutting provides a very good section. (See figs. 1 and 2, Plate XII.) This section has already been mentioned as illustrating the unconformity which separates the series from the underlying Devonian strata, but there is an added interest attached to it, for in a paper by E. J. Dunn, entitled “Glaciation in South Africa, Australia, and Tasmania,”34 one finds the following statement: — “At Cherry Gully and Silverwood railway stations, Queens- land, glacial conglomerates were noticed on a hurried trip, and considered as probably of the same age as the Ashford conglomerate, but this appears to be doubtful.” The evidence of glaciation in the Ashford conglomerate wTas disputed by David and Pittman,35 but whether glacial or not there is evidence for believing that the Permo-Carboniferous Ashford conglomerate is not of the same age as the conglomerates of Silverwood and Cherry Gully. There is every reason for regarding the horizontal Silverwood conglomerate as forming part of a series which lias been shown to lie unconformably over the steeply dipping strata of Permo-Carboniferous age. Further, the petrological nature of the boulders in the Silverwood conglomerate shovrs that they were derived from the adjacent rhyolite flows of the Eight-Mile Fault Block, which are interbedded with fossil- iferous strata of the Upper Marine stage of Permo-Carboniferous age. Again the Cherry Gully conglomerate rests upon the eroded Stanthorpe granite, which was intruded not earlier than the Upper Marine stage, as shown by dykes from the granite penetrating strata of this age in the neighbourhood. THE GEOLOGY OF THE SILVERWOOD-LUCKY VALLEY AREA. 73 All the evidence points to the Silverwood and Cherry Gnlly con- glomerates as being Mesozoic in age and very much younger than the Ashford conglomerate, but whether they are glacial or not is another matter. With regard to this aspect, the authors, after prolonged search, have not been able to find any evidence of glaciation. One does not find striated, facetted, or grooved boulders, and the arrangement of material is characteristic of near-shore lacustrine conditions. ($ee fig. 1, Plate XII.) The age of the series as a whole has not been definitely proved, but there is good reason for regarding it as equivalent to the Walloon Series of Jurassic age, although there is the possibility that its true horizon may be somewhat lower and more nearly synchronous with the Bundanba Beds. The series is separated by a marked stratigraphical break from underlying beds known to be high in the Permo-Carboniferous Series. Further, the series appears to be continuous in the field with the fossil- iferous Walloon Coal Measures at Tannymorel, some 15 miles to the east, and with the sandstones and shales of Warwick, certain horizons of which contain numerous fragments of fossil wood and which are generally regarded as being Walloon in age. Thus, although no fossils have actually been found and although the authors have not had the oppor- tunity of proving the continuity of the beds with those undoubtedly of Walloon age, they have little hesitation in referring the beds to this age. VIII. IGNEOUS ACTIVITY. [Note. — Detailed petrological descriptions of the principal rock types mentioned below will be found in Chapter XI.] The earliest evidence of igneous activity to be met with in the area is seen in the Siverwood Series, the lower portion of which is constituted iargely of andesites, spilites, and andesitic tuffs. Whether this intense and prolonged volcanic activity was terrestrial or submarine is not certain. The absence of recognisable normal marine sediments inter- bedcled with the tuffaceous material suggests the former, but the approach to the spilitic character seen in many of the andesitic flows rather favours the latter alternative. Interbedded with the radiolarian cherts which characterise the upper portion of the Silverwood Series are numerous thin beds of fine volcanic ash. The authors have shown elsewhere36 that there is good reason for ascribing these to frequent submarine outbursts of volcanic activity, and further that, as a consequence of these outbursts, the silica content of the adjacent seas was raised to the optimum condition for the growth and propagation of radiolaria, the countless remains of which are found embedded in the bands of chert and c-herty shales which alternate with the bands of tuff. This Devonian activity was not merely local, but formed part of a well-marked period of vulcanicity over a very considerable area, for Benson has shown37 that in the “ North Eastern’’ or “Tamworth” 74 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Devonian province of New South Wales volcanic activity was marked in Lower Devonian times by the eruption of great thicknesses of lavas and tuffs of a spilitic and doleritie character, while in Middle Devonian times activity was still very pronounced. At several points within the area there may be seen rather large intrusions of a dioritic type. These are undoubtedly of a later date than the Silverwood Series (although they resemble in a general way some of the interbedded andesites of the series, and more especially those which have been rejuvenated as a result of the proximity of the intruding Stanthorpe granite), but they have never vet been found intruding members of the Fault Block Series. Excellent examples of these dioritic intrusions are exposed in a railway cutting one mile north of Silverwood railway station and in the bed of Rosenthal Creek, in portion 351. In the Fault Block Series of Permo-Carboniferous age volcanic rocks form a very important feature. This has been treated at some length by the authors in an earlier publication in these proceedings,38 so that only a brief statement is required here. In four of the Fault Blocks there is associated with shallow -water marine or freshwater sediments a great development of lavas and tuffs mostly of an acid nature. These are best seen on the Rhyolite Range in the Eight-Mile Fault Block. Probably the lowest portion of this and the Tunnel Block material was laid down in a shallow sea, but the great bulk of the volcanic material of this series appears to have been extruded under terrestrial conditions. The volcanic activity of the Fault Block Series appears to have been characterised by flows and tuff's, which pro- gressively increased in basicity but decreased in amount. These outbursts appear to have been synchronous with those which characterise the Upper Marine in the Ulawarra district of New South Wales, but they present marked petrological differences, the latter being essentially potash-rich basalts and latites, while those of the Fault Block Series are much more acid and relatively poor in potash. The latest evidence of igneous activity found actually in the Silverwood-Lucky Valley area is seen in the granite which occupies a considerable portion in the south-eastern part of the area mapped. This really marks the northern margin of the very extensive mass known as the Stanthorpe granite, which lias been described in some detail by Saint-Smith.39 The erosion of this granite gives rise to topographical features quite different from those seen in other parts of the area, so that the granite forms one of the three distinct physiographic regions into which the area is naturally divided. The granite is usually coarse in grain, pink in colour, and generally resembles closely the Stanthorpe granite as developed in the type district. Near the margin one some- times meets interesting modifications, such as the granophyre and graphic granite seen at the falls in the creek bed in portion 1288, parish of Wildash. The contact metamorphism developed as a result of the intrusions of the Stanthorpe Granite into the Silverwood Series is particularly THE GEOLOGY OF THE SILVERWOOD-LUCKY VALLEY AREA. 75 interesting, for the northern line of contact is almost at right angles to the strike of the series and consequently the different metamorphic effects produced on the different members of the series may be studied. Where the granite has intruded the andesitic tuffs and andesites rO Nl 6h O, O fin ^ ^ ~ sssa cr,'“rtti§ « ‘C §-| | 2^ §•£ S 5*S r§siii?!5 % SSSsSs^f. fl SS. Ig-S a fill” $£. «o .■§ §,-§ S ««•§ a ^ aB © , 2* & a, ' ^ ^5 £ £ © I ^ S 2 1 5 §,§'£•£•£ §£ j.2 ~>£ § •S g o’s.S'gse'is g^s s © $ •5 fc £ ©•£ g.vg^'Sa IP s* SRS** AS. g svg.g s '» CC 4" >g ti ». 55 e Si C 6> „ S c<5 a, ' ■S.sgtif L,s.2i^. i&sifS 8, 5 s ^ I Sg o S S|-§&8 ■S |g §^t P-li-1 e «s e 8 a !§ 8-«a'5 8 :S'S « & 3ss g © 8s;- C^Cc^^f^fAOftnCQCQCQSQ ill:?'! I’l’lll^ I • I § &..g.g g.s “ s g g.g.g £ ^ -s.-l-l!!§,§ miiii . a Si. 58,»g >g 's vg CQOQ&a^GQ^ *, K g S ! £ fes g. s S. 05 to "« !«* e Os8S g ° S'*-'* _;-S .§ © £ £ §■ 8 - S~ © © © »«; Sft, ilttlf! QJ A< O 8- l-l £ llll? Ill-g'l s A ©, § SS ss issai,! s J J> 1 i S -i Cis Cb Cb Cb ^ b w «<■ S 8 £* •'S A© ,.S S a-S ^ sssios^ . — - s-'S'P S § w i « 8523s sill ^SSlSSiS* ii^smir?-? t^iii I lll|t|lll||l|^ si THE GEOLOGY OF THE SILVERWOOD-LUCKY VALLEY AREA. 101 The great majority of the above fossils were determined from specimens now in the collections of the Department of Geology, University of Queensland. Eurydesma Beds. — The list under this head calls for no special comment, but we would again emphasise the great number of individuals of Eurydesma cordatum and their restriction to a narrow and well- defined horizon. Wallaby Beds ( Lower Portion). — Attention has already been drawn to the preponderance of Gang'amopteris over Glossopteris in these beds. Another feature which distinguishes these from the other plant- bearing beds (Eight-Mile Creek Beds) is the presence of Glossopteris Irrowniana. A specimen of a rather unusual type of leaf collected from these beds by the authors was submitted to Dr. A. B. Walkom, who has referred it to Gangamopteris cyclopt oroides, remarking that it “has a very square apex, but is probably an abnormal frond rather than a distinct species.” This specimen (F. 1825) is now in the collection of the Queensland Geological Survey. ( See Qld. Geol. Sur. Pub. No. 270, p. 30, and Plate 4, Fig. 20.) Wallaby Beds { Upper Portion). — Procluctus is represented by one of the several species which are at present included in the one specific name “ brachythcerus .” Stropalosia. — In addition to >8. jukesii there are present forms comparable with both 8. darken and 8. gerardi, and one specimen shows features characteristic of both species. In this case the internal cast of the ventral valve is closely comparable with that of 8. clarkei as figured by Etheridge ( 8ee Geol. and Pal. of Qld. and New Guinea, Plate 13, fig. 16), while “the short, high triangular area” which Etheridge {op. cit. p. 261) regarded as one of the two most characteristic features of 8. gerardi is also evident in the same specimen. It is interesting to note that Davidson “hinted that perhaps the one might prove to be a variety of the other,” but Etheridge came to the conclusion that the two species were quite distinct. The- Spirifera include a well-marked variety of S-. tasmaniensis and numerous representatives of 8. duodecimo ostata , all of which, how- ever, lack the rib traversing the sulcus, which is a feature of the type of McCoy. In this respect the Silverwood specimens are not unique, for Etheridge points out that this does not appear to be a constant character {op. cit. p. 235), for out of eleven specimens of the species from Mount Britton, Queensland, examined by him, only two show the rib. It is an interesting fact that individuals of this species have been collected from the Eight-Mile Creek beds, which the authors regard as a somewhat higher horizon, and that they all show the median rib in the sulcus. C ondamine Beds. — This is essentially a coral fauna and has several points of interest. In the first place it affords an interesting illustration of the relationship between habit and habitat, for every species found is either a simple zaphrentoid type or else forms erect and simply branching dendroid colonies. Representing these two groups are 102 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Zaphrentis gregpriand and Trachypora wilkinsoni respectively, both typical of the Upper Marine group of the Permo-Carboniferous Series of New South Wales. Associated with T. wilkinsoni are several corals, all of a similar dendroid type, but probably representing two distinct families and at least three geuera. Unfortunately, these are represented for the most part only by casts. One coral appears to be more nearly akin to Trachypora than anything else, and may be a new species of this genus. It agrees with T. wilkinsoni in general characters and habit of growth, but differs from it in possessing much thicker walls, thus producing an even larger extent of free surface exteriorly between the calices than that noted by Etheridge in T. wilkinsoni.67 A second coral ( see Plate XVIIL, fig. 4) is referred to the genus Cladochonus, while a third is closely comparable with Monilopora nicholsoni as described by Etheridge from the Carboniferous of Mount Marmion, in North-West Australia, 68 69 for in general appearance, size, and the presence of septal flutings the two are very similar, and they agree in differing from the more typical species of the genus in their erect mode of growth. The presence of such typically Carboniferous genera as Cladochonus and Monilopora in beds which are attributed to the Upper Marine phase of the Permo-Carboniferous period is somewhat surprising, but it should be borne in mind that although Trachypora wilkinsoni, with which they are closely associated in these beds, is typical of Upper Marine strata, yet the genus is much more commonly and typically developed in Devonian and Carboniferous rocks in other parts of the world. Further, in the Permo-Carboniferous beds of the Kempsey district, N.S.W., Dun70 mentions “numerous casts of a very ramose coral which appears to be a stout variety of Cladochonus tenuicollis McCoy. ’ ’ These were associated with Zaphrentis gregoriana, arid Trachypora ivilkinsoni had previously been found in the same locality. In a hurried study of the Permo-Carboniferous fossils from Booroolq N.S.W., in the Australian Museum, the authors noted particularly specimen F. 431, where the association of a large dendroid coral with Trachypora wilkinsoni recalled vividly the similar association in the Condamine beds. Eight-Mile Creek Beds and Rhyolite Range Beds. These are thought to occupy approximately the same horizon, and may therefore be considered together. Their position beneath the volcanics of the Rhyolite Range suggests that they represent a shallow-water phase of the Condamine Beds horizon. The faunas are very different, for the predominant fossils in the Eight-Mile Block are Martiniopsis and large paheopectens, while, as we have seen, those of the Condamine Beds are corals. The flora associated with these shallow-water sediments is remark- able for its restriction to one genus of Glossopteris (G. indica) and Noeggerathiopsis Hislopi. The genus Gangamopteris appears to be quite unrepresented. [For Permo-Carboniferous fossils of the Fault Block Series, see Plates XVIII. to XX.] THE GEOLOGY OF THE S1LVERWOOD-LUCKY VALLEY AREA. 103 XIII.— ECONOMIC GEOLOGY. Within the region under consideration is the very old Lucky Valley Goldfield, of 25 square miles, which was gazetted as such on 12th March, 1869. W. B. Clarke71 “had detected by amalgamation the presence of gold in some portions of quartz from one of the veins traversing schist’7 forwarded to him by Mr. John M ’Arthur in August, 1851, and in a report dated 14th October, 1853, he writes: — “It was in that neighbourhood on what is called Lord John Swamp that fine gold, evidently of granitic character, was found more than two years ago.” C. D’Oyly H. Aplin,72 in his report of 21st July, 1869, gives an account of the Lucky Valley diggings at Elbow Gully, Lucky Valley, Frenchman’s Gully, and Duffer Gully, showing that alluvial mining had been carried on over limited areas ; he also gives a map to show the extent of the workings. He wrote that no auriferous quartz reefs had been found at the diggings, but two small leaders containing gold had been worked. With reference to ironstone, Aplin states : ‘ 1 There are three outcrops of red haematite on and in the neighbourhood of the diggings. One of them close to Oaky Creek, and about 5 miles from Lucky Valley, is fully a chain wide and is a valuable iron ore. ’ ’ Fragments of earthy carbonate of copper were found in the tailings of wash dirt taken from Lucky Valley, according to Aplin, while with respect to limestone he recorded “several outcrops of a greyish-white crystalline limestone . . . they do not appear to have any definite or uniform direction . . . they contain magnesia . . .” Tellurium is mentioned by i^plin as occurring in “small bright foliated metallic plates and scales . . . in which gold may be seen imbedded” in a quartz leader at Duffer Gully, a tributary of Lucky Valley Creek. It appears that the alluvial gold deposits were very restricted, and the field was not fortunate in furnishing an auriferous lode of any size and richness, although much prospecting has been carried on and shaft sinking has been undertaken in several places. From time to time small patches of rich quartz have been found, but nothing substantial or lasting has been forthcoming. In Mineral Gully, J mile south of Omoral railway station, it is stated that good alluvial gold was obtained many years ago. With respect to copper, on portion 107, parish of Wildash, there was a shaft sunk and work done by a local syndicate on the so-called Elbow Valley copper mine. The shaft was sunk along a much-sheared and faulted line as shown by the marble-andesite breccia and the slicken- sided andesitic material on the spoil dumps. A small jiocket of ore assaying about 13^ per cent, of copper arid 6 dwt. of gold and 20 oz. 104 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. of silver to the ton was mined, but it did not persist. Material from this development was no doubt the source of the lumps of earthy carbonate ore referred to by Aplin in 1869. Associated with the lime- stone lens at Grieve ’s quarries, in portion 660, parish of Rosenthal, there are copper carbonate stainings, while quite recently, a mile or two to the north of this, prospecting work on a copper-bearing matrix has been carried out. Here and there throughout the area, and associated with the decomposed andesitic tuffs, are surface deposits of ironstone, as has been mentioned previously. One of these surface deposits was investigated by E. C. Saint-Smith and reported upon in the Qld. Govt, Mining Journal of April, 1918. Therein it is stated that from samples obtained assay values “up to 66-2 per cent, of metallic iron” were obtained. Mr. Saint-Smith’s hope that the iron ore occurred in “a true lode formation” was unfortunately not realised by the results of the costeaning carried out in the southern part of portion 1200, parish of Wildash. The relationship between the andesitic lavas and the iron ore has already been touched upon. Between the gold and copper ores, and the granitic intrusion there seems to be some genetic relationship. Lodes, however, have not been developed to any extent, or if they were they have been almost entirely denuded, and one cannot hold out any real promise of substantial ore bodies being found in the district. The limestone lenses in the old Silverwood Series have been worked over long periods, and considerable tonnages have been quarried and burnt for lime from Grieve ’s quarries and Locke’s quarries, near Rosen- thal Creek, on portions 660 and 508 respectively. Smaller amounts have been obtained from quarries approximately 1 mile north of Silverwood station, and from near Limestone Siding some 3 miles or so further north, and from the Lucky Valley Creek area. Recently the Queensland Cement Co. has obtained a lease including the limestone in and about Reserve 49 at Elbow Valley, and it is their proposal to quarry the limestone for cement-making purposes. A tonnage of between 2,000,000 and 3,000,000 tons of suitable material, to a depth of about 100 feet beneath the surface, has been established in this one patch of limestone. Analyses of this latter material show a high lime value and an almost total absence of magnesia; this is interesting in view of Aplin ’s statement that the limestone material in these lenses contained magnesia. Owing to the proximity of this limestone to the intruding granite, contact metamorphism has resulted in the development of a lime-silicate lock on the S.E. portion of the limestone in Reserve 49. The limestone has been marmorised, not only in this reserve but also in the adjacent portions 1219, 107, 106, etc. The marble is light in colour, frequently opaque white, rather badly stained with brown limonite streaks, however. The texture is rather coarser than one would desire for monumental or decorative purposes; THE GEOLOGY OF THE SILVERWOOD-LUCKY VALLEY AREA. 105 also the jointing does not suggest one being able to obtain very large blocks. However, if the deposits were opened up it is not unlikely that the marble would improve both in respect to colour and solidity, though one would not expect the texture to get any finer. Some of the outcrops, as for example that in portion 1200, near the ironstone costeans, show less coarse marble than others. Mr. E. C. Saint-Smith,73 in reporting on these marble occurrences, stated: “The great number of joints which traverse the Elbow Valley marble occurrences will be found to prove a fatal bar to its being sufficiently extensively quarried by modern chanelling machinery when in competition with the more economically exploited deposits of about equal grade already obtained in quantity from the Ulam district. ’ ’ Some of the limestone deposits would, it is thought, produce large enough blocks when opened up properly, but in texture and appearance they would probably be inferior to other marbles just as readily available to the market. Around the edges of the limestone lenses some andesitic material has been incorporated in the limestone by the subsequent earth move- ments : also contact metamorphism has produced a certain amount of lime-silicate rock on the south-eastern portion of the limestone in Reserve 49, but the main masses of these deposits appear to be quite suitable for lime burning or cement manufacture. An average sample collected by, Mr. Saint-Smith in 1918, from portion 106, gave the followTing analyses : — Lime Loss on ignition chiefly C02 Alumina and Iron Oxide . . Insolubles Magnesia 54- 7 per cent. 43-0 per cent. 0-6 per cent. 1*2 per cent. 0-3 per cent. 99-8 per cent. XIV.— CONCLUSION. In concluding, the authors wish to express their hearty thanks to those who have in various ways assisted them in the foregoing work. They would like to thank especially Mr. J. Harward, M.A., of Warwick, whose kindly hospitality and help in the field have proved of very great value; Mr. W. S. Dun, of the Geological Survey of New South Wales, for his aid in the determination of some of the fossils and for his helpful guidance and advice ; and the Principal of the Central Technical College, Brisbane, for providing assistance for the final drafting of the block diagram, the geological map and sections. R.S. L 106 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. LITERATURE CITED. 1. Report No. x, 1853, p. 3. 2. Qld. Min. Index, 1913, p. 666. 3. Eleventh Tri-monthly Report, 1853, p. 8. 4. Report on the Auriferous Country of the Upper Condamine, Etc., p. 5. 5. Notes on the Geology of the Colony of Queensland, Q.J.G.S., 1872, p. 288 et seq. 6. Geological Features of the South-eastern Districts of the Colony of Queensland, 1879. 7. Geology and Palaeontology of Queensland and New Guinea, p. 73 et seq. 8. Qld. Geol. Surv. Pub. No. 120, p. 3. 9. Min. Res. N.S.W. No. 12, p. 10. 10. Min. Res. N.S.W. No. 14, p. 67. 11. Proc. Linn. Soc. 1909, p. 738. 12. Proc. Linn. Soc. 1913, pis. x-xiii. 13 Qld. Govt. Min. Jour. 1918, p. 354. 14. Qld. Govt. Min. Jour. 1921, p. 320. 15. Qld. Govt. Min. Jrnl. 1922, p. 359. 16. Geographical Essays, p. 346. 17. For detailed account, see “ Radiolarian Rocks in Southern Queensland” by the authors, A.A.A.S. Wellington Meeting, 1923. 18. Proc. Linn. Soc. N.S.W. 1915, vol. XL., p. 553. 19. Min. Res. N.S.W. 1908, No. 12, p. 10. 20. Benson, Proc. Linn. Soc. N.S.W., 1915, Vol. XL., Pt. 3, p. 557. 21. Benson, Rec. Geol. Sur. N.S.W., 1922, Vol. X., Pt. 2, p. 16. 22. Proc. Linn. Soc. N.S.W., XL., p. 551. 23. Proc. Linn. Soc. N.S.W., XXI., p. 553. 24. Q.J.G.S. 1899, p. 16. 25. Proc. Linn. Soc. N.S.W., XXXVIII., 1913, p. 709. 26. Qld. Geol. Sur. Pub. No. 270, PL 4, Figs. 18-22. 27. Proc. Roy. Soc. Qld., 1923, p. 109, PI. iv. 28. See Jensen, Proc. Linn. Soc. N.S.W., 1923, p. 158. 29. Mem. Geol. Sur. N.S.W., Geology 4, 1907. 30. Proc. Roy. Soc. Qld., 1923, p. 119. 31. See Andrews, Min. Res. N.S.W., 1908, No. 12, p. 10. 32. L. A. Cotton, Proc. Linn. Soc. N.S.W., XXXIV., 1909, p. 738. 33. Proc. Linn. Soc. N.S.W., 1913, p. 586-7. 34. Geol. Mag., Vol. LX., No. 703, Jan. 1923, p. 18. 35. Records Geol. Sur. N.S.W., 1898, Vol. VI., p. 80. 36. Aus. Assn. Adv. Sci., Wellington Meeting, 1923, Vo.l XVI., pp. 296-316. 37. Proc. Linn. Soc. N.S.W., XXXII., 1907. 38. Proc. Roy. Soc. Qld., 1923, p. 109. 39. Qld. Geol. Sur. Pub. No. 243, p. 39. 40. Federal Handbook B.A.A.S., 1914. THE GEOLOGY OF THE SILVER WOOD-LUCKY VALLEY AREA. 10 41. Andrews, Handbook of N.S.W. for B.A.A.S., 1914. 42. Carne. Min. Res. N.S.W., No. 14, on maps. 43. Andrews, Rec. Geol. Sur. N.S.W. , Vol. VIII., p. 113. 44. Carne. Min. Res. N.S.W., No. 14. 45. Qld. Geol. Sur., Pub. No. 243, p. 26. 46. Bryan, Proc. Roy. Soc. Qld., 1914, p. 159. 47. Bryan, Proc. Roy. Soc. Qld., 1914, p. 158. 48. Geology of New South Wales, 1922, p. 82. 49. Geology of New South Wales, 1922, p. 82. 50. Proc. Roy. Soc. N.S.W., 1921, p. 262. 51. David, Proc. Roy. Soc. N.S.W., 1911, p. 29. 52. Proc. Roy. Soc. N.S.W., 1911, p. 36. 53. Op. cit., p. 2. 54. Mem. Geol. Sur. India, Vol. XLV., Pt. 2, p. 152, PI. 35, Figs. 1 and 2. 55. Proc. Linn. Soc. N.S.W., 1920, p. 315. 56. See Dewey and Flett, Geol. Mag., 1911, p. 245. 57. See Davis, Univ. of California, Bull. Dept. Geol., Vol. II., No. 3, p. 376. 58. See Sampson, Journal of Geology, 1923, p. 592 et -seq. 59. Radiolarian Rocks in Southern Queensland, A.A.A.S., Wellington Meeting, 1923. 60. Permo-Carboniferous Volcanic Activity in Southern Queensland, Proc. Roy. Soc. Qld., 1923. 61. Radiolarian Rocks in Southern Queensland, Richards and Bryan, A.A.A.S., 1923. 62. J. S. Flett, Geol. Mag., 1919, p. 91. 63. Proc. Roy. Soc. Qld., 1923, p. 118. 64. Geol. Mag., 1923, p. 69. 65. Qld. Geol. Surv. Pub. No. 194, 1904. 66. Benson, Rec. Geol. Surv. N.S.W., Vol. X., Pt. 2, 1922, p. 19. 67. Mem. Geol. Surv. N.S.W., Palaeontology, No. 5, p. 26. 68. Bull. Geol. Sur. W. Aus., 58, 1914, p. 14. 69. Trans. Roy. Geog. Soc. S. Aus., 1916-17, p. 3, PI. XXXIX., figs. 2 and 3. 70. Rec. Geol. Surv. N.S.W., Vol. 5, p. 181. 71. Votes and Proc., Leg. Ass., N.S.W., 1853, Vol. II. 72. Votes and Proc., Leg. Ass., Qld., 1869, Vol. II., pp. 161-162. 73. Qld. Govt. Min. Journal, 1922, p. 359. 108 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. NOTE. All the photographs, with the exception of Plate X., figure 2, and Plate XII. ? figures 1 and 2, have been prepared in the Department of Geology, University of Queensland, by Mr. A. N. Falk. PLATE VII. Radiolarian Chert, showing fine banding and numerous step -faults. The coin, which is a florin, shows the size of the bands. Locality : Lord John Swamp Creek. Portion 1321, parish of Wildash. Proc. Roy. Soc. Q’land, Yol. XXXVI. I ’LATE VII. PLATE VIII. Photographs of hand-specimens of radiolarian chert, showing different degrees. of regularity in the banding. Fig. 1. — Shows bandings of different widths but of marked regularity. The lighter- coloured bands are largely tuffaceous, while the darker material, which is a typical dark -green to dark-grey chert, is almost devoid of pyroclastic material. Both types contain casts of radiolaria, but they are more numerous in the darker bands. Magnification X -6. Locality : Lord John Swamp Creek, Elbow Valley. Fig. 2. — Radiolarian chert, showing an irregularly drawn-out band of tuffaceous material. This band is composed of much coarser material than the rest of the rock. A white line has been drawn around it. Magnification X -7. Locality : Lord John Swamp Creek, Elbow Valley. Fig. 3. — Specimen of radiolarian chert with band of andesitic tuff, showing the con- siderable contortion to which the latter has been subjected. A white line has been drawn about this band. Magnification X '7. Locality i Near Sugarl oaf Mountain, The Gap. Proc. Roy. Soc. Q’land, Yol. XXXVI. Plate YIII. PLATE IX. Illustrating the limestone horizon of Devonian age, which occurs near the upper portion of the andesitic material in the Silverwood Series. Fig. 1. — Composite mass of very compact fossiliferous limestone and pyroclastic material of an andesitic nature. Magnification X ‘5. Locality : Near Morgan Park. Fig. 2. — Typical outcrop of one of the lens-shaped limestone masses of Devonian age. The enclosing rock is andesitic tuff. Locality : Oaky Creek. Prcc. Boy. Soc. Q’land, Yol. XXXVI. Plate IX. PLATE X. Fissile shales forming the uppermost part of the Silver wood Series. Fig. 1. — Shows the numerous casts of worm tracks which are typical of this horizon.. Magnification X 11 . Locality : Railway cutting near the 164-mile post. Fig. 2.— Shows the general nature of the shales and the faulting to which they have been subjected. Locality : Railway cutting near the 164 -mile post. Prgc. Roy. Soc. Q ’land, Vol. XXXVI. Plate X. 2 PLATE XI. Microphotographs of radiolarian cherts, showing the clialcecloiiic casts. Fig. 1. — Microphotograph of individual radiolarian (shown in the centre of Fig. 2). Magnification X 300. Fig. 2. — Microphotograph of radiolarian chert, showing great number of individuals. Magnification X 45. Locality : Oaky Creek, portion 2088, parish of Wildash. Fig. 3. — Banded radiolarian chert found as a pebble in a conglomerate of the Upper Marine Stage of the Permo-Carboniferous Fault Block Series. As a result of considerable pressure the rounded radiolaria have been com- pressed into lenticular shapes. Magnification 45. Locality : Eight- Mile Creek. Proc. Eoy. Soc. Q’land, Vol. XXXYJ. Plate XI. ( PLATE XII. Photograph of the basal members of the more or less horizontally disposed Walloon (Lower Jurassic) Series of freshwater sediments which rest unconformably upon- the- highly tilted rocks of the Silverwood (Devonian) Series. Locality : Railway cutting a few hundred yards on the Warwick side of the Silverwood station. Fig. 1 shows the coarse conglomeratic character of some of these basal Walloon measures. The boulders are nearly all derived from the' rhyolitic flows of the Upper Marine Stage of the Permo-Carboniferous age in the= adjacent Rhyolite Range. Fig. 2 shows the conglomerate of Fig. 1 resting upon a deep chocolate-colourecL fissile and friable shale. Prgc. Roy. Soc. Q’land, Yol. XXXYI. Plate XTI. PLATE XIII. Photographs of hand specimens, showing effects of movement on members of the Silverwood Series. Fig. 1. — Minature faulting, characteristic of banded radiolarian cherts of Silverwood Series. Magnification X ’6. Locality : Lord John Swamp Creek. Fig. 2. — Crush breccia, showing fragments of banded chert embedded in andesitic material. Magnification X -6. Locality: Near Sugarloaf Mountain,. The Gap. Fig. 3 shows the effect of disturbance of the layers of chert by currents while still in a pasty condition shortly after being laid down. A white line has been drawn along the margins of the most disturbed material. Magnifica- tion X ‘6. Locality : Lord John Swamp Creek, Elbow Valley. Fig. 4 shows telescoping of chert into andesitic material. A white line has been drawn to show the junction of the chert and the andesitic material. Magnifica- tion x -6. Locality : Near Sugarloaf Mountain, The Gap. Fig. 5 shows miniature overthrust faulting in banded cherts. Magnification X 11. Locality : Lord John Swamp Creek, Elbow Valley. Proc. Eoy. Soc. Q’land, Vol. XXXVI. Plate XIII. PLATE XIV. Photographs of hand specimens of rhyloitic flows from the Upper Marine stage- of the Permo-Carboniferous Fault Block Series. Fig. 1 illustrates fluxion structure as developed in the fine-grained pink variety. Magnification X 3. Locality : Northern slope of the Eight- Mile Range. Fig. 2 illustrates spherulitic rhyolite from The Hump, Rhyolite Range. The- spherulites have been knolinised. Magnification X '7. Fig. 3 illustrates coarse spherulitic structure. Most of the spherulitic material is now preserved as chalcedony. Magnification X '7. Locality : Northern- slope of the Eight-Mile Range. Proc. Eoy. Soc. Q’land, Yol. XXXVI. Plate XIV. PLATE XV. Photographs of Devonian Fossils from the limestone horizbn of the Silverwood Series. Fig. 1. — Part of the calyx of an undescribed rugose coral which possesses features in common with both Tryplasma and Mucophyllum. Magnification X 11. Locality : Near Limestone Siding, 6 miles south of Warwick. Fig. — Calyx and uppermost portion of stem of an undescribed rugose coral, probably identical with Fig. 1. Magnification X *8. Locality : Near Limestone Siding, 6 miles south of Warwick. Figs. 3 and 4. — Tryplasmci princeps. (Genus new to Queensland). Fig. 3 is a polished horizontal section and shows the nature of the septa. Fig. 4 is a vertical section of the same specimen, and shows the typical tabulae and radiciform process. Magnification x PI. Locality : Rosenthal Creek, 1 > miles west of Silverwood. Fig. 5. — Phillipsastrcea cf. grandis. (Genus new to Queensland.) Magnification X PI. Locality : Near Morgan Park. Fig. 6. — Syrin'/opora porteri. (Species new to Queensland.^ Magnification X P2. Locality : Near Morgan Park. Pkoc. Eoy. Soc. Q/land, Voi-. XXXVI. Plate XV. PLATE XVI. Photographs of Devonian Fossils from the limestone horizon of the Silverwood Series Fig. 1. — Favosites sp. Magnification X 11. Locality : Near Limestone Siding. Fig. 2. — Pachypora meridionalis. Magnification X -9. Locality : Near Oaky Creek Fig. 3. — Litophyllum (?) konincki. Natuial size. Locality : Near Morgan Park. Fig. 4. — Stromatoporella sp. Magnification X -9. Locality : Near Morgan Park. Proc. Roy. Soc. Q’land, Vol. XXXVI. Plate XVI. PLATE XVII, Microphotographs of Spongophyllum cf. halysitoides. Fig. 1. — Horizontal section, showing curious structure of walls, development of short septa. Magnification x 10. Fig. 2. — Vertical section, showing development of short tabul.E in middle of section. Magnification X 10. Locality : Near Limestone Siding. Figs. 1 and 2 should be compared with Spongophyllum halysitoides as figured by Etheridge fil. Plate VII., figs. 2 and 3, in the Records of the Australian Museum, Vol. XII. Pkoc. Roy. Soc. Q’land, Yol. XXXVI. Plate XVII, PLATE XVIII. Permo-Carboniferous Corals from the Upper Marine Stage, found in the Condamino Fault Block. Fig. 1. — Trachypora wilkinsoni. (Species new to Queensland). Magnification X -8. Fig. 2. — Trachypora wilkinsoni. Magnification 13. Fig. 3. — Trachypora sp. nov. This is much larger than T. wilkinsoni, a small portion of which can be seen in the upper part of the specimen. Natural size. Fig. 4. — Cladochonus ten uicollis. Magnification x ’8. Fig. 5. — (?) Monilopora nicholsoni. Magnification x ‘9. Compare with Monilopora nicholsoni as figured by Etheridge on Plate xxxix. fig. 2. Trans. Boy- Geog. Soc. S.A., 1916-17. Fig. 6. — Single corallite of (?) Monilopora nicholsoni. Magnification x 1*2. Fig. 7. — Zaphrentis gregoriana. (Species new to Queensland.) Magnification X *8. Pkoc. Eoy. Soc. Q’land, Vol. XXXVI. Plate XVIII. PLATE XIX. Fossils from the Permo-Carboniferous Fault Block Series. Fig. 1. — N ceggerathiopsis hislopi. Natural size. Fig. 2. — Gangamopteris cyclopteroides. Magnification x -6. Fig. 3. — A.viculopecten sp. Natural size. Fig. 4. — Martiniopsis sukradiata. Partly decorticated specimen. Magnification X *9. Fig. 5. — Platyschisma oculum. Natural size. Fig. 6. — Cardiomorpha cf. gryphioides. Magnification X -6. Compare with Cardio- morplia cf. gryphioides as figured by David, “ Geology of the Hunter River Coal Measures,” Plate xxxv., fig. 3. Fig. 7 . — Eurydesma cordatum, showing shape of internal cast and great thickness of shell. Magnification X *7. Localities. Figs. 1, 2, and 3 are from Wallaby Beds, portion 2073, parish of Wildash. Figs. 4 and 5 are from the Eight-Mile Creek Beds, portion 702, parish of Wilclash. Figs. 6 and 7 are from the Eurydesma Beds, portion 1292, parish of Wildash. Prcc. Eoy. Soc. Q’land, Vol. XXXVI. Plate XIX. PLATE XX. Fossils from the Permo-Carboniferous Fault Block Series. Fig. 1. — Glossopteris indica. Magnification x '9. Fig. 2. — -Aviculopecten mitchelli. Magnification x -5. Fig. 3. — Group of Productids. Magnification X -6. Fig. 4. — Strophalosia julcesii. Internal cast of dorsal valve, showing typical short septum. Magnification x 1 ‘3 Fig. 5. — Group of Bryozoa, including P rotoretepora, Fenestella, and Stenopora. Magnification x 1*6. Localities. Fig. 1. — Eight-Mile Creek Beds, portion 702, parish of Wildasli. Fig. 2. — Rhyolite Range Beds, near Limestone Siding. Figs. 3, 4, and 5. — Wallaby Beds, portion 2073, parish of Wildash. Prcc. Eoy. Soc. Q’land, Yol. XXXVI. Plate XX. GEOLOGICAL MAP OF SILVERWOOD-LUCKY VALLEY AREA. Compare with Block Diagram, page 52. For Geological Sections A and B, see pages 45 and 64. Section A is not shown on the Map, but is a generalised Section in a west-east direction from the neighbourhood of Mount Silverwood to the Condamine River. area. w00. SOT. SCO. CJ’MM, VO,,. XXXVi. SlLVERWOOD LUCKY VALLEY AREA. You XXXVI,, No, 7, 109 New Queensland Loricates. (Phylum Mollusca, Order Loricata). By A. F. Basset Hull, {Read before the Royal Society of Queensland , 26th May, 1924.) (Plato XXL) NARRATIVE. By courtesy of the Queensland Government I was invited to accompany the members of the Pan- Pacific Science Congress who were about to visit the Great Barrier Reef, travelling with them as far as Mackay. While the main party were out in the s.s. " Relief 5 5 I collected Loricates on the main- land littoral between Mackay and Townsville. As some slight recognition of the facilities afforded me by the State, the types of the new species described in this paper are deposited in the Queensland Museum, Brisbane. Arriving at Mackay on 12th September, 1923, I proceeded to Slade Point, north of the mouth of the Pioneer River. Here I spent the night at Mr. George Harrison’s seaside camp, and at dawn the next morning took advantage of the low tide and searched the rocks. The shore is muddy sand, with numerous small stones, the debris from the igneous intrusions upon the coal measures of which the country rock is formed. Along the beach these stones were waterworn and smooth or covered with small rock oysters, and only one Loricate, a dirty weed-covered Haploplax arbutum Reeve, was found. The rocky headland looked more promising, and as might have been expected, Liolophura queenslandica Pilsbry was found in immense numbers in the crevices and on the lower valves of dead oysters adhering to the surface of the rock between median and high tide marks, the bulk of the shells being greatly eroded. One specimen of Onithochiton quercinus Gould was taken in the same zone. As the tide receded I waded into the deeper rock pools and examined hundreds of loose stones, with the surprisingly poor result of a single specimen of Callistochiton antiquus Reeve, no trace of any other species being found. This was most disappointing, as the pools were deep, well sheltered, and full of algal growth. The tide was very low, and about 100 yards from the outermost fringe of rocks a sand- bank surrounded the whole point, enclosing a wide stretch of sheltered water, in which were many loose stones embedded in the muddy sand and covered with weed. Limpets and barnacles were plentiful, but although an hour was spent in diligently exploring about 400 yards along the shore under ideal conditions, favourable to the requirements of the most exigent Loricate, not another specimen was found, the net result being four species of four genera, three species being represented by single specimens only, and the other being the common Loricate of the Queensland coast, present in multitudes. R.S. — K. 110 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. I returned to Mackay and took steamer to Townsville, arriving there on the 14th. The shore in the vicinity of the town is muddy sand. There is a high bluff of porphyry at Kissing Point, but little debris at the foot and, the tide not serving, I did not collect there. On the 15th I went across to Magnetic Island, arriving there at 4 p.m. The tide was falling, and I went to the rocks at the east end of Picnic Beach. Here huge granite boulders project into the water, and the old hulk of an iron vessel lies close to the rocks. Both boulders and ship were covered with oysters nearly to high water mark. Between the wreck and an old jetty about the middle of the beach there is a large area of sandy flat, mostly exposed at low tide, and sparsely sprinkled with small stones, dead coral, and portions of an old coral reef, dead and disintegrating. I collected over this area on the 15th and on the 16th September. From the rocks and wreck I took numerous examples of Liolophura queenslandica Pilsbry, a few medium to small sized specimens of Amphitomura gemmata Blainville, and eleven specimens of Squamopleura curtisiana Smith. The two former species were evidently sedentary, wedged into crevices or between or inside the valves of the thickly clustering oysters, their girdles distorted to fit the irregularities of their environment ; while the lastnamed species were evidently mobile, looking fresh and alert and moving when touched. On the flat area I found numerous specimens of Ischnochiton luticolens Hull and Haploplax arbutum Beeve, the former on rough granite fragments, dead coral or smooth stones ; the latter principally on smooth stones and old bottles. On a piece of rough granite I found an Acanthochiton belonging to the same beautifully figured and sculptured group as A. granostriatus Pilsbry ; this shell I am describing in this paper, the name, A. complanatus , referring to the remarkable flattened pustules with which it is sculptured. Seven specimens were taken, all within an area of a few square yards. On the 17th September I left Townsville for Bowen. Here I was fortunate in meeting Mr. E. R. Rainford, who kindly placed himself and his excellent boat at my disposal for two days, and without whose assistance I would have been unable to do any collecting, there being no boats for hire in Port Denison. On the 18th we went to North Head, where a long sand- bank stretches from the rocky headland, on which the lighthouse is erected towards the shore. The sides of this bank were thickly strewn with debris from the granitic and dioritic rocks of the vicinity. Here we found Ischnochiton luticolens and Haploplax arbutum in moderate quantity, both on the under side and at the edge of insertion in the sand of loose stones, besides in several cases on living or dead shells of the large Pinna. Three specimens of another Ischnochiton, which I am describing and naming I. distigmatus, were found. Mr. Rainford found the first of a new species of Lucilina, which I have much pleasure in naming after him ; about twenty specimens of this shell were taken in all. Several specimens of Callistochiton antiquius Reeve, two Acanthochiton complanatus, and two A. variabilis Ad. & Angas, were also taken. The main rocks were cursorily inspected, but only a single Liolophura queenslandica was found. On the following day we sailed across to Stone Island and anchored off a long sandspit on the eastern side of the island. This spit was covered with rocky debris, piled up on both NEW QUEENSLAND LORICATES. Ill sides but covered at high tide. Along the shore of the bay, which terminated in a rocky low headland, there was a wide stretch of coral sand with scattered stones and shingle partly embedded in it. Nearly every stone examined on the flat yielded a single I. luticolens or H. arbutum, and on the under side of some smooth stones embedded with loose pieces of coral in the sand I found a most interesting new genus and species of Plaxiphoridce, a quite unlooked-for discovery, no species of the family having been recorded from tropical Queensland. On searching the low headland, Mr. Rainford found a fine example of Callistochiton granifer and three specimens of Rhyssoplax particolor, two species recently described by me from Palm Islands and Caloundra, Q., respectively. As the rising tide drove us from the flat we examined the rocks on the spit, taking several fine examples of Amphitomum gemmata Blv. SPECIES OF LORICATES COLLECTED. 1. ISCHNOCHITON LUTICOLENS. Ischnochiton luticolens Hull, Aust. Zool., iii, 159 (1823). Plentiful at Magnetic Island, Townsville, North Head and Stone Island (Port Denison). (Type locality : Port Curtis, Q.) 2. Ischnochiton distiomatus, n. sp. (Plate XXI, Fig. 1.) Shell small, elongated oval, not carinated. Colour purplish-brown when alive ; much of the purple disappears when the shell is dry ; a dark purple spot on each side of the jugum near the posterior margin of valve iv. The sculpture of the whole shell is finely granulose, graduated in quincunx. The lateral areas of the median valves are raised, but there is no differentiation in the sculpture. Posterior valve larger than the anterior, with mucro rounded, slightly in front of the middle, posterior slope convex. Girdle densely packed with small, elongated, lozenge-shaped scales directed backward and outward, not striated ; on the underside there are closely packed spicules, radiating outwards from the suturaljnargins to the outer edge of the girdle. Interior, bluish-white ; anterior valve interiorly grooved and with about 9 rudimentary slits, median valves 1-1 (in one case there are two obscure slits on one side), posterior valve crenulated but unslit. Dimensions. — 8 x 4J mm. Station. — On the under side of small stones embedded in the sand between median and low water mark. Locality. — North Head, Port Denison, Q. Material. — Three examples. Remarks. — This shell is easily distinguishable from I. luticolens Hull, with which it is associated, by the absence of any marked differentiation in the sculpture on the lateral areas ; the uniform colouration, and the shape of 112 PROCEEDINGS OP TPIE ROYAL SOCIETY OP QUEENSLAND. the posterior valve, which in I. luticolens has the mucro considerably in front of the middle and the posterior slope concave. It differs from /. gabrieli Hull in the obscurely slit median valves. It appears to be a degenerate species, the insertion plates and slitting becoming obsolete, and the girdle scales approaching the elongated spiculose character of those of the Lepidopleuridce. 3. Haploplax arbutum. Chiton arbutum Reeve, Conch. Icon, iv, pi. xxiv’ fig. 162 (1847). Plentiful at Magnetic Island, Townsville ; North Head and Stone Island, Port Denison ; rare at Slade Point, Mackay, Q. 4. Callistochiton antiquus. Chiton antiquus Reeve, Conch. Icon, iv, pi. xxv, fig. 169 (1847). Chiton antiquus Reeve; Smith, Zool. Coll. “Alert,” 79 (1884) Port Molle (Coppinger). Not found at Magnetic Island ; eight examples at North Head, Port Denison ; one example at Slade Point, Mackay. 5. Callistochiton granifer. Callistochiton granifer Hull, Aust. Zool., iii, 161 (1923). A single specimen of this rare species was taken by Mr. Hainford at Stone Island, Port Denison. 6. Acanthochiton complanatu.3, n. sp. (Plate XXI, Figs. 2, 2a, 2b, 2c.) Shell medium, elevated, carina ted, side- slopes convex. Colour variable ; the type shell has a creamy ground, greenish on the jugum, and each valve is strongly marked with irregular zigzag lines in greenish-black, as depicted in figure 2. Other shells are suffused with emerald green or brown. The sculpture is similar on all valves, consisting of numerous almost circular flat-topped pustules, increasing in size towards the margins. Anterior valve rounded, not trilobed, with apex almost smooth. Median valves ; lateral areas not differentiated, no diagonal ; jugum broad, V-shaped, with the pustules very small and crowded. Posterior valve small, rounded, the pustules smaller and tending to a more regular pattern than on the other valves ; mucro prominent, posterior, straight behind. Girdle densely clothed with short transparent spicules, seven large tufts of long spicules on each side, opposite the sutures, and four smaller tufts in front of the anterior valve. Interior bluish-white ; anterior valve with five slits, median valves 1-1 slit, and posterior valve with two slits. Dimensions. — 15 x 8 mm. 113 NEW QUEENSLAND LORI CATES. Station. — Under small stones embedded in coral sand. Locality. — Magnetic Island, Townsville (type) ; North Head and Stone Island, Port Denison. Material. — Magnetic Island, 7, Stone Island, 3, North Head, 6 examples. Remarks. — Differs from A. granostriatus Pilsbry in the flattened, circular pustules. Differs from A. shirleyi Ashby in the regular, more closely packed and flattened pustules, and the elevated nature of the shell. Smith (Zool. Coll. Alert, 1884) recorded a shell from Port Molle under the name of A. asbestoides. Iredale informs me that he has seen the shell in question, which was collected by Dr. Coppinger of the “ Alert,” and he considers it probably conspecific with the shell above described. It certainly was not A. asbestoides, the type locality of which is Flinders Island, Bass Strait. 7. Acanthochiton variabilis. Hanleya variabilis Ad. & Ang., P.Z.S., 1864, p. 194. Nine examples were taken at Stone Island, and two at North Head, Port Denison. Aebilamma, n. gen. Shells of small dimensions and tropical distribution, having Plaxiphorid structure, but primitive ornament, and a girdle covered with minute scales, bearing sparse corneous processes, and with spiculose margins. Colour resembling oxidised bronze. 8. Type : Aebilamma primordia, n. sp. (Plate XXI, Fig. 4.) Shell very small for the family, broadly ovate, not elevated, but carinated ; side-slopes convex. Colour olive-green, unevenly flecked with reddish-brown and blotched with dark green. The sculpture of the whole shell is uniformly granulose, the grains somewhat irregular in shape and size, but all high and rounded. Anterior valve broad and high, apex not beaked, plana te. Median valves ; lateral areas distinctly raised, not differentiated ; beaks prominent on valves ii to iv ; grains on the jugum smaller and more crowded. Posterior valve ; mucro small, posterior, not prominent ; post-mucronal area indicated by a slightly raised rib. Girdle wide, covered with minute, elongate, rounded scales, and having a few corneous tufts opposite the sutures ; margin spiculose, the under side densely minutely scaly. Interior blue, paler on the edges and the sutural laminae ; anterior valve with 8 slits, median valves 1-1 slit, posterior valve unslit ; callus wide, smooth ; eaves projecting ; sinus broad. Dimensions. — 7x5 mm. (type). Largest example taken 11 x 7 mm. The measurements are of dried specimens, the girdle contracting to less than one-third of the width of the living shell. 114 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Station. — On the under side of smooth stones embedded in coral sand and debris, between median and low water mark. Locality. — Stone Island, Port Denison, Queensland. Material. — Seven examples. Remarks. — The discovery of this shell is of great interest, the nearest recorded occurrence of a Plaxiphorid in Queensland being at Caloundra, 50 miles north of the Brisbane River, while Port Denison is nearly 700 miles further north. The southern record is of Poneroplax ( Plaxiphora ) pceteliana Thiele, a large species which attains a length of 80 mm. on the coast of New South Wales. As indicated by the specific name proposed, A. primordia is primitive in its sculpture, and resembles in this respect Plaxiphora parva Nierstrasz (Beit, zur Kenntnis der Fauna von Sud-Afrika, Zool. Jahrb. 1906, 501), and P. indica Thiele (Rev. des Systems der Chitonen, Chun. Zool., 1909, 23). The former shell is from Mozambique ; its dimensions are 5 x 3J mm., and the colour “ white, red, or grey, flecked with black.” The figure indicates a more elongate shell than A. primordia, but the granulose sculpture is similar, although less uniform. Thiele’s shell is from Ceylon ; his type is 8 x 5| mm., colour “ brown flecked with green.” I have an example from the type locality, 15 mm. in length, in which the sculpture tends to radial arrangement, but otherwise is in agree- ment with that of A. primordia. These two species, placed by their authors in the genus Plaxiphora, appear to fall into the same genus as the Queensland shell, both having scaly girdles. The Plaxiphoridce are primarily of Antarctic distribution, being found from the Falkland Islands to the sub- Antarctic islands of New Zealand, where they assume heroic proportions. The com- pletion of the trinity — African, Indian, and Australian — of tropical degenerate forms, furnishes another remarkable item of evidence in support of the theory of the northward migration and synchronous degeneration of Antarctic fauna. Vide my paper on the occurrence of the crested penguin in Australia, Rec. Aust. Mus. xii, 79 (1918). 9. Rhyssoplax [particolor. Rhyssoplax particolor Hull, Aust. Zool., iii, 165 (1923). Three examples collected by Mr. Rainford at Stone Island, Port Denison. 10. Squamopleura curtisiana. Chiton ( Ischnochiton ) curtisianus Smith, Zool. Coll. “ Alert," 78 (1884). Eleven examples of this shell were collected at Magnetic Island from the wreck of an iron steamer. Oysters were clustered thickly along the sides of this wreck, and in the early morning the Squamopleuras were crawling about the oysters. 11. Amphitomura gemmata. Chiton gemmatus Blainville, Diet. Sci. Nat. xxxvi, 544 (1825). Amphitomura gemmata Blainville, Iredale, P.Z.S., 1914, 669. Amphitomuras were fairly common on the wreck and rocks at Magnetic Island, but small shells predominated. On a rocky spit at Stone Island some NEW QUEENSLAND LORICATES. 115 large examples were taken, and also some of remarkably fine condition as regards sculpture. The examination of these and numerous examples from other parts of the Queensland coast, Torres Strait, and North Western Australia leads me to conclude that there are many varieties, subspecies, or even valid species of the genus. Ashby has suggested the varietal name of queenslandica for the shell from Dunk Island, Queensland1, but as there are several names available if Queensland shells are to be separated from Torres Strait and Western Australian shells, I prefer to await a more critical examination of material before definitely accepting or rejecting Ashby’s suggested name. It may be noted that the foot of the animal when alive is orange-red. 12. LlOLOPHURA QUEENSLANDICA. Liolophura gaimardi queenslandica Pilsbry, Proc. Acad. Nat. Sei. Phil ad., 1894, 87. Liolophura gaimardi, var. queenslandica Pilsbry, Hedley, A.A.A.S., 1909, 352. Liolophura queenslandica Pilsbry, Hull, Aust. Zool., iii, 199 (1923). This is the common shell of the southern coast of Queensland, which I first found associating with the preceding species on the rocks of Facing Island, Port Curtis. It occupies the same zone, and accommodates itself to the irregularities of the rock surface, oysters, and other encrusting growths, in the same manner as the Amphitomura. The foot of Liolophura is generally of a pale yellow colour. I found it plentiful on the rocks and wreck at Magnetic Island, rather scarce at Port Denison (but further search will probably result in its discovery in quantity), and very plentiful at Slade Point, Mackay. 13. Onithochiton quercinus. Onithochiton quercinus Gould, Proc. Bost. Soc. Nat. Hist., ii, 142 (1846), New South Wales. This is one of the commonest shells inhabiting the surface of the rocks between high and median tide marks on the coast of New South Wales and Southern Queensland. I found only a single example at Point Slade, Mackay, and none at either Port Denison or Magnetic Island, 14. Lucilina Rainfordiana, n. sp. (Plate XXI, Fig. 3.) Shell medium, elevated, carinated, side-slopes very slightly convex. Colour very variable, brilliant when alive, in most instances fading to dull brown, red or green when dry. The type shell is pale brown, flecked with red, especially on the jugum ; valves ii and viii greenish- black with a light stripe on the jugum ; central areas of valves iii to vii greenish. Anterior valve with eleven radiating grooves, the interspaces filled by 4-5 imbricating pustules with the apices directed backward. Ocelli visible all over the valve. Median valves with lateral areas strongly differentiated, covered with three radiating rows of large flattened imbricating pustules, the central row smaller than the two outer ones. Central areas having 10-12 high curved 1 Journ. and Proc. Roy. Soc. W.A., viii, 30 (1921-2). 116 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. riblets, the bases corresponding with the anterior row of pustules on the lateral areas, and curving inwards to the jugum ; ocelli visible on all the valves, mostly scattered along the grooves in the lateral areas. Posterior valve ; mucro very prominent, posterior, nearly straight behind ; ante-mucronal area similar to central areas of median valves, with ocelli visible between the riblets ; post-mucronal area densely covered with small rounded pustules. Girdle wide, spongy, densely covered with minute, short, chaffy spicules. Interior bluish-white. Anterior valve with 8 slits, teeth very finely pectinated and exteriorly crenulated ; median valves 1-1 slit ; posterior valve with 12-13 slits, teeth short, broad, coarsely pectinated and externally crenulated. Sinus broad, finely pectinated. Dimensions. — Type, 15 x 1\ mm. Some examples were taken measuring 20 x 12 mm. Station. — On the under side or at the edge of insertion in sand of stones and dead coral, in coral sand below low-water mark. Locality. — North Head, Port. Denison. Material. — About 20 examples. Remarks. — This is a very striking shell, brilliant and variably coloured when alive, and having a very strong sculpture, easily distinguishable from that of any other member of the genus. In the juvenile shell the apical half of the anterior valve is smooth, and there are only three concentric rows of flattened pustules ; the sculpture of the lateral areas of the median valves consists of two rows of pustules only, obscure in the very young shells, with a smooth area between. In the senile shell the rows of pustules on the anterior valve and the lateral areas of the median valves show a tendency to increase by division towards the margin. I have much pleasure in associating this shell with Mr. E. H. Rainford, of Bowen, whose kindly assistance enabled me to obtain this and other interesting Loricates at Port Denison. EXPLANATION OF PLATE XXI. Fig. 1. — Isclinochiton distigmatus, n. sp. Fig. 2. — Acanthochiton complanatus, n. sp. .Colour pattern. Fig. 2a. — Acanthochiton complanatus. Anterior valve. Fig. 2b. — Acanthochiton complanatus. Median valve. Fig. 2c. — Acanthochiton complanatus. Posterior valve. Fig. 3. — Lucilina rainfordiana, n. sp. Fig. 4. — Aerilamma primordia, n. sp. Plate XXI. Proc. Roy. Soc. Q’land, Vol. XXXVI. [Face page 116.] Vol. XXXVI, No. 8. 11T The Artesian Waters of Queensland. By P. C. Tibbits, A.M.I.E. (Aust.), Engineer for Boring, Queensland Irrigation Commission. (Plates XXII.-XXIV.) ( Bead before the Royal Society of Queensland , 28th August , 1924.) The discovery of artesian water in Queensland dates back to 1887 — about thirty-seven years ago — when a bore was sunk in what was known as “Cotton Bush” paddock, on Thurrulgoona Station, in the Cunna- mulla district ; later in the same year* another bore was sunk in Barcal- dine. Both yielded a large flow of water. At this time practically nothing was known of the sources of artesian waters. The judgment shown by those responsible for fixing the location of the two bore sites mentioned deserves special recognition ; providentially, or otherwise, the fact remains to-day that no better sites could have been selected if all the data of to-daj^ had been available. Both these bores, though so far apart, are situated in areas which have perhaps been now more fully developed by boring than any other parts of Queensland. I mention these facts to show that there was something* more than “pot luck” operating, and I have little doubt that those responsible must have had a fair knowledge of the origin of artesian water. Since that time much has been written regarding this subject,, a good deal of controversy has taken place amongst eminent geologists, and their views are to this day divided as to its source or origin. Like most other subjects, it appears to me that whilst the researches by geologists have given some data for determining certain conclusions,. I am not convinced that the lines which they have followed, and from which their deductions have been derived, are sound enough to enable them to solve this important matter. I therefore propose to give some new data which may throw fresh light on the subject, and which may materially assist geologists to treat the matter from a geologically structural point of view. I have prepared maps and sections from ascertained data which I feel certain confirm the theory held by me, viz, that the source of supply is entirely meteoric, and not plutonie or a combination of both. Hitherto the area in which artesian water was discovered has been described as the “Great Australian Basin.” This assumption of there being only one basin is, to my idea, distinctly erroneous. From evidence obtained from what is known as the Hydraulic Survey of this State I think I can claim to show that no less than eight distinct basins have been determined. These basins, or, in other words, underground lenti- cular water-bearing beds, have their source at the foot of the various ranges from which only a small percentage of the rainfall run-olf is absorbed in the saturated area leading into the different strata in r.s. — Ij. 118 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. which artesian water is temporarily impounded. What is meant by ‘‘temporarily impounded” is that the water, being in slow movement towards its natural outlets or safety valves, its impounding can only be described as temporary, for it certainly has proved a diminishing quan- tity, due to overtapping. No greater evidence could be provided to prove that there are leakages in the strata, from which we derive our artesian supplies, than the existence of the ‘'mound springs” in the vicinity of Lake Eyre. A study of the map showing the various basins will, I think, convey the impression that each and every basin has a distinct outlet. Evidence of Mound Springs. The evidence afforded by the existence of these natural mound springs in the desert area around Lake Eyre in South Australia gives ample proof that in ages past the saturated porous strata were fully charged under an enormous static head, and it is only reasonable to assume that this water would tend to rise to its own natural intake level, less the loss of head due to friction. The altitude of the natural surface level of the desert referred to being only a few feet above sea level, it is quite natural that the impounded water would break through in the weakest places. The effect of this happening was the beginning of the formation of these huge mounds. It is generally known that all artesian bore water contains a fairly large percentage of solids in solution, but the composition of these solids varies considerably in different localities. The formation of these South Australian mounds consists of the consti- tuents in the water having solidified when in contact with' the air. This is quite a common occurrence, and this action going on for ages has resulted in mounds being formed to a height of about 280 feet, which stand out prominently in the middle of the desert country. When the maximum static head had been reached no more building up took place, and the water remained stationary. At this prehistoric period there was no other outlet for the water from this particular source. All other distinct areas have their natural outlets, but it is not proposed at this juncture to go into details in connection with them, except to mention that the map produced will give a good idea of them. Again referring to the mound springs, we find that, since the. sinking of artesian bores, the effect has been to gradually lower the static head of the water, and in consequence the water level in the highest mounds gradually receded and the tops of the mounds developed a slightly concave formation until the water became completely sealed down. With the lessening of the static head, due to the multiplicity of bores, these mounds do not now-a-days rise more than about 40 feet above the natural surface. Unfortunately, sufficient data in the way of levels and pressures are not available of bores in the adjoining State of South Australia to enable it to be definitely stated whether the water level in these mound springs agrees with the isopotentials of the bores sunk within the area in Queensland from which the escape takes place. Typical mound springs in varying stages of formation were photo- graphed on the spot. THE ARTESIAN WATERS OF QUEENSLAND. 119 Is Origin Meteoric or Plutonic? The opinion has at times been expressed that the source of supply of artesian water is plutonic, i.e., that the water is derived from deep- .seated rocks and is driven into the overlying sandstones by contraction during cooling off of an almost molten mass. In the same category can be placed the theories that the water was impounded in the sandstones when laid down, and sealed off in them by the over-lying shale, and that it derives its pressure from either pent up gasses accumulated since deposition or from the excessive heat of these strata thereby causing expansion. For some time the field staff engaged on the Hydraulic Survey has been endeavouring to collect information which would throw light on this subject by measuring the actual temperatures at the bottom of various bores with a recording thermometer suspended from piano wire. Naturally these results, especially in corrosive areas, can only be obtained in a few cases on account of the difficulty in lowering the instrument down any but well-cased bores. Even where the bottom temperature cannot actually be determined, a fairly reliable opinion can be formed as to whether the temperature is abnormal by comparing the flowq depth, and surface temperature with corresponding figures for a nearby bore, at which the bottom and subsurface temperatures have been recorded. It will readily be seen that if the temperature at the bottom of any hole be abnormally high the temperature of the flow at the surface will also be higher than expected from a comparison with surrounding bores, as the only loss is that caused by conduction through the casing to the cooler rocks, and this is a fairly constant quantity for any given area. So far it has always been possible to estimate the bottom temperature within 5 degrees before recording it, and occasionally within 1 degree. The onty bores which may really be considered anomalies, viz., Alva and Saxby No. 8, have surface temperatures which are far and away above anything in the locality. From this ,it seems that it may be justly assumed that future subsurface measurements of existing bores will not disclose any outstanding features such as Alva and Saxby, but, of course, it is not claimed that no future new bores will resemble Alva and Saxby. The usual rise in temperature for the Northern basins varies fairly evenly from as low as 1 degree Fah. in 45 feet, in the southern portion, to about 1 degree Fah. in 18 feet in the north, while at Saxby it is 1 in 13 feet and at Alva as high as 1 in 11 feet. The general accepted average for the earth is given as 1 in 60 feet, therefore it must be admitted that the plutonic theory supporters can at least point to excessive heat over the whole of our basins. In investigating these temperatures it is thought that it might be possible to indicate definite centres of heat, but so far, with the two exceptions given, the change has been very gradual and similar to that of the isopotential lines, but not in the same direction. Now, if it is granted that heat is the source of pressure we would surely expect that an excessive temperature would give at least an 120 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. appreciable rise in pressure. This is not the case. At Saxby the iso- potential, as far as can be calculated in this corrosive area, is not deflected, and we can say generally that the subsurface temperature seems to bear no relation to the pressure in the beds. At Alva it might appear that we did obtain an unusual pressure, since the estimate of the probable pressure was 20 feet less than the actual. At the same time the estimate was based on the assumption that we would encounter a bed of a similar porosity to those at another Trust Bore further down stream, whereas we probabty struck a free sandstone and obtained 1,150,000 gallons per day Avith 140 feet head, at between 1,100 and 1,200 feet, with a temperature of 159-J degrees Fah. The chief objection to all phases of the plutonic theory is that the isopotential lines strongly suggest a source of supply towards the eastern side of the basins. If confined gases were the source of pressure wre Avould expect the direction of flow in the beds to be from all sides, towards any heavily tapped area, and that therefore equal pressure lines wmuld be practically closed figures, whereas they are roughly parallel in a north-west and south-east direction. Keplenishent Theory Sound. Another important feature bearing on this point is the occurrence of mound spring areas near Lake Eyre in South Australia and other somewhat similar groups, such as Eulo, Hungerford, and Springvale, in Queensland. It is not generally conceded that these, and particularly those around Lake Eyre, act as permanent outlets. This, to my mind, disposes of any theory which does not admit of replenishment of the beds at the present time, since if there is no water entering the beds these leakages must have released all surplus pressure ages ago, as there are mounds in all stages of existence in the vicinity of Lake Eyre. If there AA^ere no evidence of mature mounds in which Avater had receded wre might look on the Lake Eyre leakage as of comparatively recent origin, but since all trace of a mound that had ceased to operate would become obliterated by the sand dunes, it is reasonable to assume that this leakage has been going on for practically a geological age. We have seen the effect of the supply of thirty-seven years’ draught, varying from^ nil to 300,000,000 gallons per day, so that it is almost certain that a few million gallons per day for tens of thousands of years must have released all surplus pressure, unless the supply had been continually replenished, whereas, in the early days of boring, flowing supplies were- obtained almost level with the outcrops of the porous beds at the foot of the dividing ranges to the east. The natural mound springs referred to indicate that they were, and are still, to a certain extent, the safety valves, and consequently must be admitted as one of the natural escapes of the artesian Avater. It is held by some that the fact that we still obtain large supplies disposes of the theories which do not admit of replenish- ment, but this is hardly the ease. The total draught at present on the beds is equal to only about two points of rain over all the basins of THE ARTESIAN WATERS OP QUEENSLAND. 121 Queensland. This can easily be accounted for by expansion of con fined water in the beds due to heat. What appears to be a serious objection to theories which support a replenishment by plutonic water is the fact that there is practically no evidence of similar supplies outside the artesian (Cretaceous) area. If these igneous rocks which, generally speaking, come to the surface outside the basins give off water beneath the sandbeds, we might reasonably expect them to do so outside the areas. Even though the amount to replenish the present draught need only be about two points of rain per annum getting into the saturated area within the intakes, we would at least expect some evidence of it in various valleys outside the area. Objections TO' Meteoric Theory. The chief objection to the meteoric theory seems to be the possibility of obtaining the requisite supplies by soakage along the crest of the main dividing ranges. Now we can trace water-beds with moderate certainty at gradually decreasing depths as we approach the ranges, until we come to localities, such as Aramac, where flowing supplies are obtained at a few feet (about 10 feet) in an area abounding with small springs. Beyond this we have continuous sandy areas in which shallow non- flowing supplies can be obtained. It appears to me that it matters very little if in some areas the age of the sandy surface formation differs slightly from that of the water-beds, provided these come sufficiently near the surface to draw on the ground waters which we know exist on the “desert” country on the range. Since the width of these sandy areas is about 40 miles or so in places, and the greatest width of the basins less than 600 miles, we can see that less than half an inch of rainfall reaching the beds at the intake above Aramac would supply the bores in a direct line between it and Lake Eyre, since the average draught on the basins only represents two points of rain over the intake areas. This is by no means an excessive amount to be disposed of by ‘ ‘ ground ’ ’ water, and therefore large areas could be unproductive without affecting the meteoric theory. Another apparent anomaly is the fact that we have not yet been able to detect a measurable fluctuation in supply due to varying seasons. (The researches of Slichter on the motion of underground waters throw some light on this subject.) For sands similar to ours, and under some- what similar slopes to those that occur near our ranges, the measured velocities in the beds vary only a few feet per day, say, about a quarter of a mile a year. Now the conditions which must obtain at the intake, if the meteoric theory is to hold, is that the rain soaks into the sand and percolates along its base as ground water, until it enters the water- beds proper^ which are sealed off by impervious overlying beds. It will be seen at a glance that no measurable pressure can develop further upstream than the edge of the impervious beds, as the water is merely soaking through the sand-beds which are not saturated to the top. The whole annual rainfall on these beds represent only 5 or 6 feet of saturated sand, and even if this fell all in one storm the general saturation level 122 PROCEEDINGS OP THE ROYAL SOCIETY OF QUEENSLAND. would rise only 6 feet at most, and the water would then percolate at the rate of a quarter of a mile a year towards the water-beds from an area several miles in width. It is therefore not wonderful nor surprising that varying yearly rainfalls produce little fluctuation in supply. In my opinion there is ample area of porous sands and sandstones, in those portions of the “ intake” which I have visited, to provide the amount of ground water required to support the meteoric theory. Of course it is only to be expected that in such a large area there would be groups of bores which did not conform absolutely to any theory yet advanced, but I view these as merely exceptions which by no means, upset the whole theory. For instance, there is a large loop in the isopotential on Barealdine-Aramac area, which might be due to the fact that the sands beyond Aramae are rather lower than elsewhere, or that there is a submerged range separating the intakes, and consequently the pressure in the beds varies accordingly. The Kynuna area also^ behaves rather erratically, since there we find another “loop” in which the diminution has been very large until the bores became non-flowing. Taken by itself, this area has given very even results, and it appears to suggest a sand bed of imperfect connection with the other beds. Lenses. Boring records, taken as a whole, do not suggest a series of indepen- dent lenticular beds of sand. Of course there are cases, such as Kynuna, where it would appear as if the main water-bed was in imperfect connec- tion with the surrounding beds, since its rate of diminution was high,, although the area has not been unduly overtapped. Generally speaking, I would feel inclined to consider some of the upper beds as more or less lenticular layers in imperfect connection with the main beds. This would explain phenomena such as we get on the Warrego, where bores tapping the upper beds originally gave large supplies and high pressures, and are still flowing, but with small supplies and a very low rate of diminution. These could be explained by the above hypothesis, for with an imperfect connection between the several beds, once the upper bed was developed, most of the available pressure wrould be used up driving or forcing the water into the top beds. The first bores to tap this would get big supplies, because there would be very little friction between the beds, and they would be tapping an accumulated supply. However, as the beds were developed or depleted the pressure would keep on falling (as at Offham No. 1 on the south-west) until it developed only a few feet of head. Nevertheless, although the present pressure of the beds is low we would not expect the supply to completely give out sooner than in the main beds, as it really represents a restricted outlet from them. Offham No. 1 Bore in 1912 had a pressure of 19 feet head, and it is still flowing, although surrounding bores with big pressures are losing as much as 7 feet per annum. I do not suggest that all upper beds are of this nature, but the explanation would apply to such bores as Offham, Elmina in the south-west, and Kynuna in the central-west. THE ARTESIAN WATERS OF QUEENSLAND. 123; Quite likely most of the upper beds have lower pressures, because they derive their pressure supply from ground water at lower altitudes, i.e,,. their cover rocks outcrop further west than those overlying the main beds. Basin Depths. Although generally agreeing with the map recently published by the Chief Government Geologist, giving the depths at which the main artesian supplies may be obtained throughout the basins, I would hesi- tate to delete the references in older maps to the Palaeozoic rocks south- west of Thargomindah. These were put on the original map as the result of reports of gold ‘ ‘ f ossickers ’ ’ working in the Grey Range area,, and do not seem to be out of place when compared with identified rocks at Tibooburra and Broken Hill. Of course it is beyond question,, especially on the evidence of Patchawarra Bore in South Australia, that there is a very deep area near Lake Yamma Yamma, but T would feel more inclined to predict shallower depths in the south-west corner of this State, near Stokes Range. If this be so, the country in that vicinity would warrant development by sinking bores, and this is well worthy of serious consideration. Critics will probably raise the point that, while representatives of different departments entertain a variety of opinions on various phases of the artesian question, it is unwise to enforce restrictive legislation. Practically all authorities now admit — what has long been insisted upon by the Water Supply Department — that it has been proven beyond doubt that there is a serious falling off in available supply, due to the multiplication of bores. It is, therefore,, wise from a national standpoint to forbid the sinking of any bore to water country which can be supplied by an existing bore. Definition of Basins. The accompanying map (Plate XXII.) has been prepared with the object of defining the boundaries of the different artesian water-bearing beds, which are numbered numerically from 1 to 8 (shown in block squares) . Basin No. 1 takes in that area lying south-easterly from the Gulf of Carpentaria. This area, especially in the vicinity of Richmond, has been largely tapped, with the result that during the last ten years the rate of diminution has greatly increased. This decrease, it is contended, is due to the multiplication of bores. Evidence has been obtained by means of the hydraulic survey from which isopotentials have been prepared, and which indicate the trend of underground water in the direction of the Gulf of Carpentaria, and it seems quite reasonable to believe this water has a natural outlet in that direction. Basin No. 2 is in the central- west, bounded on the north and east by defined ranges, and on the west by a limestone belt. Isopotentials in this area are quite distinct from No. 1 ; therefore the evidence is that the sources of supply are also distinct. The outlet for this basin has not been definitely defined, for the reason that, the isopotentials indicate 124 PROCEEDINGS OP THE ROYAL SOCIETY OF QUEENSLAND. that it is in the Federal Territory, of which nothing is known. It will be seen that the eastern dividing range of this area is shown somewhat broken, but it is quite likely the same formation is continuous, though in places partially, or wholly, submerged. Basin No. 3 has its centre in the Longreach district with its intake along the western boundary of the Great Dividing Range. The characteristic of this basin is that towards the north, viz., in the Aramac district, artesian wafer is obtained at a very shallow depth, viz., about 10 feet, but the bed dips very fast in a westerly and south-westerly direction. Isopotentials point to the outlet being coincident with the confluence of the Thomson River and is absorbed in Lake Yamma Yamma. Basin No. 4. — This particular basin has only been slightly tapped in the vicinity of Lake Buchanan, and, unlike the remainder of the basins, apparently has its intake on the eastern side of the Great dividing Range. The hydraulic survey has not been completed in respect to this area. Sufficient information has, however, been obtained to enable it to be said that artesian water exists. My opinion is that the water from, this basin finds its way to the Pacific Ocean between Bowen and Towns- ville, and perhaps forms the delta of the Burdekin River. Basin No. 5. — Little is known of this area for the reason that com- paratively few bores have been sunk in it. It is known that artesian water is only obtained at depths of upwards of 4,000 ft. The natural outlet for this basin would be in the direction of Lake Eyre in South Australia. Basin No. 6. — This is perhaps the most interesting basin in the whole State, owing to the fact that it is possible to define seven distinct basins within the basin proper. Perhaps I would be more accurate if I separated these basins instead of treating them as one, but I will endeavour to describe the existence of these beds and will treat them as shown on the map. It will be seen that there is a small range separating each of the main watercourses flowing southerly. In places these ranges are practi- cally submerged, but so far as separating the artesian waters from one another they appear to be effectual. This is borne out by the fact that the flow obtained and static head attained are quite dissimilar and bear no relation to one another, therefore it -is concluded that these ranges abut on the Main Range to the north, and the catchment within each is confined to the respective basins. I am unable to give any definite information as to the outlet of Basin No. 6 and its sub-basins, as it extends into New South Wales. Basin No. 7. — This basin has its source from the dividing range north of Roma and Dalby, or from practically the same as Basin No. 6, but there is in this case likewise a difference in head. The direction of flow points to the outlet being south-westerly and extending into New South Wales. Basin No. 8. — This basin differs somewhat from Basins Nos. 6 and 7 in that the isopotentials show that the source of supply is entirely derived THE ARTESIAN WATERS OF QUEENSLAND. 125 from the highlands to the east. Ai group of five bores has been sunk by the Government in the Weengailon Settlement, from which definite data have been collated showing that the hydraulic gradient is from north- east to south-west, and not north to south-westerly as in the case of Basins 6 and 7. The outlet for this basin likewise lies beyond our border. Having defined the areas of the respective basins and in'takes, I think it will be readily admitted that the indications point to the theory that the whole of our artesian waters are primarily derived from rainfall run-off from Queensland ranges, and not from the run-off from high mountains in New Guinea, as suggested by Mr. Dunstan. When it is considered that it took many centuries to give us our maximum storage, it is not wonderful that the draught would in time exceed the intake •supply. The fact that each and every basin shows a gradual but sure decrease in both flow and head proves that the storage supply is limited and compatible with the capacity of the porous strata in which it accumu- lated, and each and every basin appears to have a natural safety valve, viz., its outlet. Considering that there are over 1,300 artesian and about 3,000 subartesian bores sunk within the basins shown, it can easily be comprehended what an enormous underground supply would be neces- sary to yield the present output of over 300,000,000 gallons per day without affecting the supply. The time has long passed when the equili- brium was reached, viz., that point at which the output equalled the supply, and the vital question now is, and for some time past has been, how to deal with this question from an economic point of view. There are many other important features that could be dealt with in this paper, but in view of the fact that the evidence given suggests a theory of origin entirely different from any other theory advanced, perhaps I have provided enough data to form the subject of a lengthy debate by our able geologists and. engineers. Bores in any One Basin Have Equal Isopotential Value. The section of a tract of country illustrated by Plate XXIII. is taken on a line running generally in a south-westerly direction from a point north of Roma. It connects nine bores which derive their water supply from the same source, which is practically proved by the regular isopotentials obtained. It will be seen that these isopotentials represent a true hydraulic gradient from the highest to the lowest altitudes. The hydraulic gradient of each of these bores has been obtained by actual pressure tests and, when plotted, these give a true unbroken line indicating that the intake source is at an elevation somewhat higher than the flowdng level. The surface levels and isopotentials are given for each bore, so that it is clearly shown that a relative connection exists, and in lieu of the pressures in the different bores producing a horizontal plane, i.e., the water rising to its intake level, a sloping plane is produced, the reason for this being attributed to loss due to friction in the passage of water in the porous strata from the higher to the lower levels. I have adopted this section line to indicate generally that a group of bores sunk 126 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. within any defined area can be immediately associated with a particular source of intake, and I unhesitatingly state that it is quite possible to reliably estimate the pressure of any new bore to be sunk within an area of which a series of pressure tests have been previously obtained. To reliably estimate the flow of any particular bore in a defined area is* quite another matter, as it depends to a great extent upon the density of the water-bearing strata, and this varies largely. When a flow is obtained in a stratum of dense porosity, the period over which it takes to record the maximum pressure extends for many hours, whereas the maximum pressure in a porous stratum is usually obtained in about one to two hours ; therefore when a bore is sunk on the outskirts of one of the artesian basins, there is always a risk of striking a more dense stratum with a consequent smaller flow, with the same relative head, but it takes longer to acquire it. The diminution rate of flow from the bores referred to in this section, or bores which have their source on the highlands of this* particular area, shows an uniform loss of 1-9 per cent, per annum, which is regarded as the second lowest in the State, but this rate cannot be expected to remain constant if the multiplicity of bores continues. After carefully investigating all the available data collated in respect to the flowing life of artesian bores, I have no compunction in saying that in less than sixty years artesian bores in this area will be a thing of the past, and long before that time it will be realised what this national loss means; therefore a warning should be heeded and restrictive measures taken to fully utilise the waters already flowing before permitting the sinking of further bores, as assuredly every additional bore accentuates the rate of diminution. It is not contended that because the water will cease to flow there will be none available. I do not consider the available supplies will ever be exhausted by artificial pumping, for the reason that only such quantities as are actually required will be pumped, and this will never equal the rate of intake nor exhaust the already stored supplies ; therefore the economic residual value will be enormous and perhaps more fully cherished than the present services. The geological section gives a fair idea of the positions and dimensions of the lenticular water-bearing beds, from Roma south-westerly, and it is drawn from the actual logs of bores sunk, i.e., from Roma downwards — that portion from Roma upwards is assumed only. From a glance it will be seen that the lenses dip towards the lower altitude, and are not horizontal nor isolated as suggested by Mr. Dunstan in his recent lecture (Roy. Soc. Qld., 24th Nov., 1923). I do not hold with his view as to the supply lying dormant in lenses, for the reason that I fail to see what element would operate to give the actual static head recorded in each bore. In passing, I might mention that if the water-bearing beds and other strata, as described in the geological section referred to, be continued in a northerly direction to the foot of the dividing range, it appears to me that the prospects of a storage of oil in this locality would be hopeless, for the reason that as the strata dips there are no reservoirs in which oil is likely to accumulate. The fact that large flows of gas are encountered in this area does not, to THE ARTESIAN WATERS OF QUEENSLAND. 127 my mind, indicate that flowing oil will be obtained. The same inflam- mable gas was struck in all the bores referred to in this section, and in one case (Whyenbah) the drilling tools, weighing approximately fifteen hundredweight, were carried up the borehole about 300 ft. and at a faster rate than the cable could be coiled round the spool. This will give a fair idea of the pressure of gas encountered, so if there is oil in this region at all it seems more likely that it would be carried down by this gas pressure to the lower regions and not held up in the Roma district. I merely record this as an opinion only, and as it has some bearing on the intake area of this basin it may be interesting from a geological point of view. Bores of Varying Isopotentials in Different Basins. The section shown in Plate XXIV. is taken practically as a straight line from Thargomindah due easterly to Boondandilla, and it shows that between each river there is a defined elevation, or partly submerged range. These ranges are more pronounced in a northerly direction, and are shown in perspective, and undeniably indicate the formation of the boundaries of the seven artesian basins (as shown). The relative poten- tials of bores in each of these basins are/ distinctly different, and tend to confirm my opinion that to prepare an isopotential map by connecting up bores in the different basins would be totally misleading and inaccurate. It has always been my contention that isopotentials can only be plotted from data collected from bores with the same source of supply; therefore on the evidence prepared on this section I think it must be admitted that isopotentials extending right across the section area would be worthless. Bach and every defined basin has quite an unequal potential value, indicating that the sources of supply to each vary in altitude, but as the outcropping porous sandstones forming the intake areas have not, in all cases, been connected by levels, it cannot be said that their relative altitudes agree with the isopotentials. The fact remains that bores sunk in each basin produce a variation in the potential, and to my mind this clearly indicates separate beds and 'also separate intakes. The rates of diminution of flows also vary, e.g., in the Warrego basin the diminution is on an average as high as 3-J per cent, per annum, whilst in the area between the Paroo and Bulloo Rivers the average diminution rate is only about 1-6 per cent. This is the lowest rate of diminution in the State, but it might be pointed out that this particular area has only been slightly tapped, and, as the multiplication of bores takes place, then will the rate of diminution increase, so that in this area the flowing life may not exceed between sixty and seventy years. The altitudes and potentials of bores sunk along this section line are given as they exist, and the occurrence of the dividing ranges between each is so pronounced that the conclusions already stated have, I consider, been logically arrived at. I have dealt only with the south-western portion of the State, so far as providing sections is concerned, because 128 PROCEEDINGS OF TPIE ROYAL SOCIETY OF QUEENSLAND. to embrace the whole State, or the individual basins, would take up more time than I have for this work. Perhaps additional data to that now available will be required before this can be done, but the sections prepared will open up fresh matter for debate, and that is my principal reason for preparing this paper. It is pointed out that all bores sunk to the same water-bearing beds within the seven different basins extending across this section line give a relative static head, whereas to ^connect each by isopotentials the result is a mass of contortions ; therefore it seems self-evident that only bores within a defined basin can be reliably plotted for isopotentials. Economic Value. Having dealt with the origin or source of supply of the artesian water, it might be interesting to give some idea of the way in which this water is utilised. A map (exhibited) shows twenty-seven Trust bores, the areas watered by each, and the location of the drains therefrom. Each of these drains carries water all the year round and affords stock and domestic water supplies, and is more serviceable than a river, for the reason that it is accessible throughout its entire length, and when consolidated is not a death-trap to stock. To show the whole number of bore trusts already dealt with by the Government throughout the State would entail the preparation of a very large map. I have, therefore, confined my remarks to the south-western portion of the State only, from which a good idea can be obtained as to the great service rendered. About 4,000,000 acres have already been treated in this way by the Government, the water distributed in upwards of 2,600 miles of bore channels, and all charges for interest and redemption on capital expended are met by those benefited, and the general taxpayer bears no burden whatever in connection with loan expenditure. When it is considered that these Trust bore areas represent only about 2 per cent, of the total artesian area within the State, a very good idea can be obtained as to the real national value of artesian water for the development of bur pastoral lands. One has only to reflect for a moment to realise what a national calamity is in store for Australia when this great asset disappears. That it will, in time, disappear is certain. Conclusion. In this paper I have set down a new theory of the origin of our artesian waters. I contend — (i.) That our artesian waters are entirely derived from rainfall run-off within the State of Queensland. (ii.) That eight independent basins have been actually defined. (iii.) That evidence has been given that some, at least, of the above basins are divided into sub-basins by partially submerged ranges. (iv.) That it is not practical to connect all bores by isopotentials. Artesian Water.- MAP of Queensland showing Approximate Locations of Water-bearing Beds, DEDUCED FROM Varying Potentials. 108 m2 miles . AM./. Engc§ (Aust.) Enc/c for Boring, Q '/anc/ irr/q. Comm. c. Roy. Soc. Q’land., XXXVI.— Plate XXHI. 11. XXXVI.— Plate XXIV. : V..L XXXVI. No. 9 129 On a New Species of Pandanus from North Queensland. By fouNT Prof. U. Martelli (Florence, Italy). (Plate XXV.) ( Communicated by C. T. White, Government Botanist, Brisbane, 29th September, 1924.) Pandanus ( Keura ) Whitei Martelli, n. sp. Folia ultra 150 cent, longa, 8 cent, lata e basi ad apicem sensim attenuata, acuminata, coriacea, in parte basilari crasse coriacea, ibique, in pagina superiori, late concavo-canaliculata, deinsuper anguste et profunde canaliculata, apicem versus plana, e basi usque ad medium a duobus plicis lateralibus percursa, deinde evanescentibus, utrinque longitudinaliter crebre et minute venosa sed, basin versus, venis evanescentibus, propterea laevia, ibique sub vitro minutissime impresso- punctulata apparet ; marginibus inermibus, praeter quam in tertia inferiori parte, brevi tractu, tantum dentatis ; dentibus brevibus, subulatis, erectis; costa media in pagina inferiori, vix prominenti, inermis. Syncarpium non vidi (sed puto magnum et globoso-oblongum, similliinum ab illo P. Cookii). Phalanges ponderosae, obovatae, in medio incrassatae, vertice subtruncato-convexiusculae, 7 cent, longae, 6-6-| cent, diam., dimidia parte, inferiori fibrosae, superiori convexae, nec punctulatae, nec scabriusculae, irregulariter longitudinaliter profunde sulcatae ; loculis 9-10, subaequalibus, magnis, a sulcis plus minusve profundis separatis; longitudinaliter ± valide rugoso-costulatis, superne, loculis prominentibus, irregulariter rotundato-pyramidatis et obscure penta- gonis, acute costulatis, interdum etiam longitudinaliter rugosis. Stigma parvulum, horizontale, planum, hyppocrepiforma, ad verticem unusqui- que loculi situm. Endocarpium osseum, 3J cent, spissum et totam latitudinem phalangis occupans, superne rotundatum sed prope ad circumferiantiam adsurgens. Mesocarpium superum, 2 cent, spissum; cavernis latis, fibroso-medullosis. Habitat : Low savannah country near Townsville, North Queensland. Legit C. T. White, 3rd January, 1922 (Herbarium Martelli). During my study of several collections of Pandanus I have noticed that these interesting plants have been rather neglected by collectors of the Australian flora. It is difficult, indeed, to preserve complete specimens of them for the Herbarium, as it is necessary to gather ripe fruits and male inflorescences. These latter are as yet only known of 130 PROCEEDINGS OP THE ROYAL SOCIETY OP QUEENSLAND. a very few species, and for this reason w.e want at present to have at least some phalanges for correct specific determinations. I therefore take this opportunity of calling the attention of Australian botanists, as well as those of other countries, to the importance of these plants in order to be able, in a short time, to complete the Monograph published in 1900 by Prof. Warburg in Pflanzenreich (Engler), which appears to be too incomplete a work for the needs of systematic botany. As regards the Australian species of Pandanus, one notices that those belonging to the “Section Keura” predominate; in fact, among the eleven species known up to the present, nine belong to this section, one to the “ Acrostigma, ’ ’ and one is incertae sedis because known only by male flowers. The Australian species of “Keura” on account of the different appearance of the phalanges of the different species may be grouped into the following types: — The first is represented by the P. tectorius form ; the second by that shown by the P. Coohii, P. Dam- manii , and P. Whitei group, a characteristic Australian type in the shape of its phalanges and which grow in North Queensland. The third type is represented by the very interesting P. spiralis P. Brown ( non alior ), which grows in the Palmerston region (Port Darwin), Northern. Territory. Proc. Roy. Soc. Q’land, Yol. XXXYI. Plate XXY. [ Face page 1 30. j Vol. XXXVI., No. 10. 131 Radiolarian Jaspers in the Brisbane Schist Series. By Professor H. C. Richards, D.Sc., and W. H. Bryan, M.Sc., University of Queensland. (Plate XXYI.) (Read before the Royal Society of Queensland, 29th September, 1924.) (I.) INTRODUCTION. The following account makes no pretensions to completeness, but aims at putting on record the discovery of radiolarian jaspers in the Brisbane Schist series, and at pointing out the possible bearing which the discovery may have on that vexed question, the age of the Brisbane Schists. Some few weeks ago the authors visited the Fernvale district, on the Brisbane Valley Railway, 12 miles to the north-west of Ipswich. About three-quarters of a mile north-east of Fernvale railway station a small watercourse crosses the intersection of two roads, one of which runs in a south-westerly direction towards Fernvale railway station, the other being at right angles to this. On the western bank of the small watercourse there outcrops a series of red and chocolate coloured rocks dipping almost vertically and striking west-north-west. These form the westernmost members of the Brisbane Schist series at this point. The outcrop is made up of red jaspers alternating with a very decomposed rock which appears to be an altered andesitic tuff. On closer examina- tion, certain of the jasper bands were found to contain radiolaria, while one horizon in particular showed the casts of these organisms in great numbers. The discovery of radiolaria was in itself an interesting one, but the fact that no recognisable fossils had ever before been found in , the Brisbane Schists made it also an important one. (II.) THE BRISBANE SCHIST SERIES.1 The Brisbane Schist series includes most of the old unfossiliferou's rocks in Southern Queensland, and the series is continuous into north- eastern New South Wales. It is undoubtedly of very great thickness, and its members display considerable lithological diversity. While in the type district, i.e., in the neighbourhood of Brisbane, the prevailing rocks merit the term “schist,7 7 yet much of the Brisbane Schist so-called is not accurately described by that name, for although the whole series has been subject to heavy folding and crumpling, schistosity has not been produced in all cases. Petrologically, the series varies from the glaucophane schists and allied rocks described by Jensen2 at one extreme 1 For a more detailed description, see Richards, Proc. Inst, of Engineers, Aust., Brisbane Divn., 1922. 2 A.A.A.S., 1909, p. 263. 132 PROCEEDINGS OP THE ROYAL SOCIETY OF QUEENSLAND. to almost unaltered shales at the other. This lack of lithological uniformity, together with the great thickness of the series, has led to the suggestion that it represents considerably more than one period of deposition. J ensen, for instance, Was inclined to regard the series “as a Pakeozoic complex of deep-sea deposits ranging from, perhaps, Pre- Cambrian to Devonian or Carboniferous in age.”3 This diversity in lithological character is in part responsible for the very different ages which have been assigned to the series as a whole by different geologists, for these estimates were, in the absence of fossils, based largely upon lithological resemblances to series of known age in New South Wales and Victoria. Thus the series has been tentatively assigned in whole or in part to each of the following periods: — Archaean (David)4, Pre- Cambrian (Wearne)5, Pre-Cambrian to Carboniferous (Jensen)6,. Ordovician (David7, Dunstan8, Richards9), Silurian (Rands10), Devonian (A. C. Gregory11, Dunstan12), and Permo-Carboniferous (Jack)13. The official view of the Queensland Geological Survey at the present time is that the series is doubtfully Ordovician. An important advance in our knowledge of the Brisbane Schists was the recognition by Mr. B. Dunstan, Chief Government Geologist of Queensland, of the natural division of the series into three belts.14 In the lowest and most eastern of these “there is a general widespread occurrence of phosphates such as apatite, turquoise, and wavellite.” The middle belt is commonly manganiferous, while the upper or western belt is characterised by the presence of serpentines and limestones. “This sequence is so persistent that in a locality where a series occurs one or both of the others are also to be found, conspicuously so in areas in the Rockhampton and Gladstone districts and about Kilkivan and Ipswich.” The uppermost belt Dunstan regards tentatively as of Devonian age. In addition to the phosphatic, manganiferous, and serpentinous belts, the authors have frequently considered the signi- ficance and use of what appeared to them to be a definite belt of ferru- ginous quartzites and jasperoid rocks which they have observed at many points in the field. 3 Proc. Roy. Soe., Qld., vol. xxiii., p. 154. 4 Fed. H Wk B.A.A.S., 1914, p. 259. 6 A.A.A.S., 1911, p. 124. * Op. cit. 7 Contribution to A.A.A.S., 1913. 8 Appendix B., Harrap ?s School Geography of Queensland. 9 Op. cit. 10 Qld. Geol. Sur. Pub. No. 34, p. 1. 11 Report Geol. Features S.E., Qld., 1879. 12 Manganese, Art. 8 in Industrial Minerals. 13 QH. Geol. Sur. Map, 1892. 14 Op. cit., p. 250. RADIOLARIAN JASPERS IN BRISBANE SCHIST SERIES. 333 (III.) THE RADIOLARIAN JASPERS. Since their discovery, field work has proved that the radiolarian jaspers are much thicker and more extensive than was at first realised. If one leaves the Fernvale railway station and proceeds in a north- easterly direction, he passes over the open rolling country of the Esk (Upper Triassic) series of lacustrine sediments for about three-quarters of a mile, when the Mesozoic strata are abruptly terminated and are succeeded (in portion 65) by the Brisbane Schist series. The line of junction is emphasised by the change in the topographical nature of the country, the resistant members of the Brisbane Schist series weather- ing into rugged hills. The boundary between the two series forms (so far as it was followed) a straight line and, as Cameron has pointed out, it is probably a faulted junction, being a continuation of the great West Ipswich fault which is usually associated with his name. The first member of the Brisbane Schist series met with, if one continues the section in a north-easterly direction, belongs to the upper- most belt of Dunstan and is a handsome serpentine which shows great evidence of having participated in strong folding movements and is almost schistose in places. It appears to be striking about west-north- west and to be dipping almost vertically. These observations are con- firmed by the rocks which immediately succeed and underlie the serpentine. They are made up of bands of red brittle jaspers, varying from a few inches to as many feet in width, which alternate with very decomposed chocolate-coloured andesitic tuffs. The jasper bands strike west-north-west and dip very steeply (almost vertically) to the south- south-west. Many of the jasper bands contain fossil racliolaria, while some horizons are crowded with traces of the minute organisms. Con- tinuing the section to the north-east one crosses a steep ridge made up largely of massive ferruginous quartzites and jasperoids. Radiolaria were not found in all these, but the lithological uniformity, when con- sidered with the occasional discovery of radiolaria, left no reasonable doubt that they all formed part of one great formation. The last point in the section where radiolarian jaspers were found was in portion 129, near the bend in the road where Stinking Gully closely approaches it. The section, if continued to the Brisbane River, affords no further information, as the rocks are covered by recent river alluvium and river gravels. Whether the formation continues to the east of the river has not yet been determined, but the section as described passes over just half a mile of steeply dipping jaspers and interbedded andesitic rocks. A reasonable estimate gives a thickness of 2,500 feet. The great thickness of the radiolarian jaspers as seen in the Fern- vale section proved that this was more than merely a local development, and might therefore be expected to occur in other localities. This expectation, was tested by making a section in the Pine Mountain area about 6 miles south-east from Fernvale. This area is an inlier of the Brisbane Schist series in the Mesozoic coal measures, and is practically on a continuation of the line of strike of the Fernvale section. R.S. — M. 134 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. This section proved in all essentials, as was expected, very similar to that at Fernvale. A section across the inlier from sonth-west to north-east showed first the schistose and steeply dipping serpentines striking about north-west, and these were followed by the ironstones, jasperoids, and ferruginous quartzites of Pine Mountain, many of which were found to contain radiolaria. Associated with the ironstone in this section is a crystalline limestone which unfortunately proved unfossiliferous. That jaspers lithologically very similar to those at Fernvale and Pine Mountain are widely developed in Queensland is undoubted, but whether they all contain radiolaria and whether they all represent the one series, are problems which have not yet been investigated. Mr. L. C. Ball, B.E., who has a very extensive knowledge of Queensland formations as seen in the field, wrote some years ago concerning the Pine Mountain jasperoids : — ■ ‘ Such a rock is of universal occurrence throughout Queensland in connection with the iron and manganese deposits in the older rocks.”15 The authors, too, have frequently met this rock type in the field in Southern Queensland usually associated with the Brisbane Schist series. (IV.) THE AGE OF THE RADIOL ARI AN JASPERS. The radiolarian jaspers do not possess in themselves direct evidence as to their age, but some hint may be obtained by comparing them with rocks of the same type of known age. Other things being equal, it is most reasonable to assign tentatively to them the age of the nearest rocks which closely resemble them. Another clue might be found by inquiring into the age of the rocks with which the jaspers are most closely associated in the field. These are the Pine Mountain and Fernvale serpentines. Applying the first method, we find that the nearest rocks of known age which are closely comparable are to be found many miles to the south in the Woolomin Series of Lower Devonian age in the New England district. Here the parallel is remarkably close, both as to (a) lithological types, (&) presence of radiolaria, (c) relationship to a great fault line, and (d) association with serpentine. In both cases a serpentine line striking north-westerly follows a great fault which divides a newer series on the west from an older series on the east. In each case the older series lying to the east of, and closely associated with the serpentine, is made up for the most part of ‘ ‘ slaty siliceous rocks, reddish banded cherts, and red jaspers in which the traces of radiolaria can be faintly discerned. . . .”16 Other radiolarian jaspers are known in Eastern Australia from several periods, but as these are all much further away from the area under consideration, and as none of them afford nearly such a close 15 Qld. Geol. Sur. Pub. No. 194, 1904, p. 55. 16 Benson, Proe. Linn, Soc., N.S.W., 1913, p. 494, RADIOLARIAN JASPERS In BRISBANE SCHIST SERIES. 135 parallel as the one just mentioned, one must conclude that the evidence, such as it is, favours the correlation of the Fernvale-Pine Mountain jaspers with the Woolomin Series of Lower Devonian age. If we can assign an age to the serpentine we can fix an upper limit to the age of the jaspers. Benson17 has shown that the great majority of the serpentine bodies of Eastern Australia occur as steep sill-like masses associated with folded palaeozoic strata and having much the same trend, and has suggested that they may be coeval and that all may be associated with the same great folding movement. The Tasmanian serpentines seem to be Devonian in age, and Benson at first assigned the Great Serpentine Belt of New South Wales to a Late Devonian age, but afterwards referred it to Late Carboniferous times. Dunstan correlates the Pine Mountain serpentine with that of Kilkivan to the north and with the Canoona-Cawarral serpentines still further to the north, all of which he regards as probably of Devonian age. When all the evidence is considered, it seems safe to conclude that the radiolarian jaspers are older than Upper Devonian. On the other hand, as they form part of the uppermost of Dunstan ’s three divisions of the Brisbane Schist series, and as the tentative acceptance of an Ordovician age for the whole series is at present based on the phosphatic minerals found in the lowermost belt, they are probably younger than the Ordovician. Possibly the arrangement may be as follows: — 1. Radiolaria have been described from a new locality in Queensland. This makes necessary an addition to the list of radiolarian rocks in Australia which forms Table II. in the authors7 “Radiolarian Rocks in Southern Queensland.”18 2. The radiolarian jaspers and associated rocks of the Fernvale-Pine Mountain area bear a close resemblance in many respects to the Woolomin Series of New England. 3. A tentative correlation of the area under discussion with the Woolomin series would give the former a Lower Devonian age. 4. A Lower Devonian age for the Fernvale-Pine Mountain area is quite in harmony with the views: — (a) That the serpentine belts of Southern Queensland are of Upper Devonian age and ( b ) that the lowest or phosphatic members of the Brisbane Schist series are of Ordovician age. Brisbane Schist Series. Serpentines Radiolarian Jaspers Manganiferous Schists Phosphatic Schists . . Upper Devonian Lower Devonian ? Silurian Ordovician (V.) CONCLUSIONS. 17 A.A.A.S., 1911, p. 104. 18 A.A.A.S., vol. 16, pp. 296-316, 1923. 136 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Plate XXYI. Figure 1. — Microphotograph of radiolarian jasper, showing c-halcedonic casts of radio lari a set in red jasper. Note serrated edges of casts. Magnification X 100. Locality: Three-quarters of a mile north-east of Fernvale Railway Station. Figure 2. — Mierophotograph of the same specimen showing great numbers of radiolaria. Magnification X 30. Compare this plate with Plate XI. of the authors’ “Geology of the Silverwood- Lucky ValleY Area. ” these Proceedings, Yol. XXXVI., 1924. Proc. Roy. Soc. Q'land, Vol. XXX VI. Plate XXYI. Figure 2. [Face page 136.] Vol. XXXVI., No. 11. 137 A New Rotifer of the Melicertan Family. By W. R. Golledge. (Plate XXVII.) [Read before the Royal Society of Queensland , 27th October, 1924.) I have the pleasure of introducing an interesting addition to the Rotiferse of Queensland, and, in all probability, to that of the world. The discovery is really due to a Brisbane lady, Miss Bursdorff, so I have named it Melicerta bursdorff ce. The Rotifers, or Wheel Animalcules, are a group of minute, highly - organised, and interesting animals. Hudson and Gosse divided them into four sections, and in the first the Rhizota, or those which are rooted or fixed in one place, the subject of our paper, finds its appropriate place. It is attached to the filamentous leaves of a Vtricularia, one of our well- known insect-eating plants, and is only to be found towards the end of autumn. It measures, when extended, one-tenth of an inch, so that it is just visible to the naked eye. The large, four-leaved corona, spreading like a white lily, is its most conspicuous feature. It sways back and forward, a perfect living flower. The double lines of cilia, bordering the lobes, are always in active operation, sending a swift current along the channel between bearing food to the mouth. Under the microscope the tiny jaws of the malleo-ramate type are seen to open, and close, once every second. Just above the stomach are two gastric glands, and behind lies the ovary. The eggs, large, oval, and dark in colour, are extruded through the anus and retained by a mucous attachment around the mouth of the aperture until they are hatched, as many as five being observed clustered on one specimen. Prom the anus tapers a long transparent foot, which suddenly contracts near the base to a firm tube which is imbedded in the substance of the plant. The animal is of a composite character. The head is decidedly Melicertan. The foot is that of a Floscule, and structurally it resembles the Megalotrocha. In the light of accumulating facts it is evident that a revision of the group is needed, but the pronounced head renders the Melicerta the most suitable family for its reception at present. I forwarded specimens to Mr. J. Shepherd, of Victoria, a gentleman who has given special attention to Rotifers, and he has supplied me with a good deal of information, and a table showing the points of likeness and disagreement with, the other genera of the group. Thus the new animal agrees with two generic characters in Lacinularia and Megalotrocha, and with only one in Melicerta — viz., four lobes. The absence of the pellet- forming organs and secreting tubes leads it away from Melicerta and places it nearer Megalotrocha. The specific characters may be given: Corona large and four-lobed, dorsal gap minute, ventral antenna very 138 A NEW ROTIFER OF THE MELlCERTAN FAMILY. small, dorsal antennas possibly absent, occurring singly, length 24 mm., breadth -37, corona -5 mm. wide, -7 mm. deep. Four flame cells showing through corona on either side. Pellet or gelatinous tube absent, foot contracted suddenly to short terminal rod. ■ — - Corona. Dorsal Gap. Antennae V entral Antennae Dorsal. Clusters with Tubes. Clusters without . Tubes. Opaque Warts. Melicerta . . Four-lobed Wide Obvious . . Minute . . None None None Lacinularia Heart- shaped, oblique, long Minute . . Absent or minute Adhering gelatinous tubes None None Megalotrocha Kidney- shaped Minute . . Absent or minute None None Four New Species Four equal lobes Minute . . Minute . . Absent . . None None None The best-known of the Group is Melicerta ringens, which has been known for over a hundred years. It forms an interesting microscopic object to all lovers of aquatic life. Judge Bedwell wrote a lengthy description, and in Pludson and Gosse ’s great work it is most beautifully delineated and charmingly described. Another member of the family is Melicerta tubicolaria. I got a specimen in the Enoggera Reservoir ten years ago, but have not seen one since. It is enclosed in a gelatinous cell, and is distinguished by two long antennai. In a pool at Goodna I found another form, which I think has never been described before, and I named it Melicerta coloniensis. It was distinguished by each member possessing two brilliant red eyes. They gleamed like a pair of rubies, and formed an exquisite sight under dark ground illumination. Another species found is Melicerta conifera. Its tubes are conical in shape, and young ones frequently attach themselves to the tubes of other specimens for support. Another variety I have seen occasionally is Melicerta janus. It forms a tube of oval pellets, but these are formed in the intestines. These examples will show you how much the subject of this paper differs from the other known members, and that it forms a very interesting addition to the group. REFERENCES. Hudson, C. T., and Gosse, P. H., 1889. “The Eotatoria or Wheel Animalcules,” 2 vols., London. Proc. Roy. Soc. Q’land, Yol. XXXVI. Plate XXVII, [ Face page 138.] You XXXVI., No. 12. 139 Geological Notes on the Country Between Childers and Munduberra. By H. I. Jensen, D.Sc. (Read before the Royal Society of Queensland, 24 th November, 19,24.) Introductory. — The writer had recently an opportunity to visit this district for the fifth time, but on this occasion was able to traverse a large area and connect up what were formerly only isolated observations. Viewing the geological map accompanying Mr. Dunstan’s Mineral Index, we find the Maryborough beds marked Cretaceous and including the Burrum coal measures. We see the Tiaro coal measures marked Ipswich and intruded by granitic and rhyolitic bosses as at Mount Bopple and Biggenden Bluff. The metamorphic sedimentary strata west of the Tiaro coal measures as far as the Dawson basin are coloured Gympie and are extensively intruded by granites. On my first visit to this area in 1918, I was under the impression that the country between Biggenden and Mount Shamrock was Devonian on account of the high degree of metamorphism of the beds. That impression was wrong, and as far as I can see at present, after several tours, the position as regards stratigraphy is the following: — The Maryborough beds are Cretaceous or Upper Jurassic ; it is for palaeontological research to determine. They are not affected by the acid intrusives, volcanic, hypabyssal, and plutonic that affect the Tiaro coal measures. The Tiaro coal measures are, in the writer’s opinion, of Ipswich age, the same as the Esk beds. They are intruded by dykes of acid igneous rocks and as a result of the intrusions they are much folded and the coals are high in fixed carbon. The slates, schists, crystalline limestones, and quartzites found west of the Mesozoic beds are in the main Permo-Carboniferous (Lower Bowen, which may be Upper Carboniferous). They are rich in large crinoidal remains. The limestones of Curra, Tamaree, Gigoomgan, Biggenden, Paradise, Marule, Gin Gin, and other places probably belong to this series. The sedimentary beds are steeply inclined; often, as at Marule, nearly vertical, giving one the impression of high geological antiquity. They are intruded about Degilbo, Chowey Creek, Mount HO PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. Shamrock, and elsewhere by basic sills and dykes (diabases and gabbroid rocks) and by still later granites, porphyries, and rhyolites. The basic intrusive® are often much metamorphosed, especially west of Gayndah. But it is highly probable that, considering the whole belt from Childers to Camboon, we are dealing with several series of basic intrusives of vastly different geological ages. In my opinion the sedimentary rocks met with between Gayndah and Mundubbera are older than the Lower Bowen. At Phillpot Creek we get limestones' . cherts, jaspers, andesites, quartzites, and schists inter- bedded, intruded by basic diabases which themselves are highly meta- morphosed. The limestones are of a blue or dark colour, contain abundant coral remains and a few small crinoid stems. They are, in my opinion, Devonian. Volcanics are interbedded with the sedimentary rocks. Strike. — The strike of the Permo-Carboniferous series at Biggen- den, Booyal, Mount Shamrock, and in this belt is approximately north-south. The strike of the series met with west of Gayndah, as at Phillpot Creek, is between N.N.E. and N.E., which in Central Queens- land generally is a very old strike direction. On personal experience I regard N.E. strikes predominating as pointing in general to Siluro- Devonian age, N. strikes as Upper Palaeozoic, and N.N.W. strikes as Mesozoic and Tertiary. I may he right or wrong, but the study is worth following up. Childers to Dallarnil , east of Doomby, that is east of the Childers red soil area, we have the poor soils of the Tiaro coal measures with a flora of wallum, tea tree, white gum ( Eucalyptus hcemastoma var. micrantha), honeysuckle ( Banksia ), and similar poor country genera. The Childers red soil area, formerly all scrub (the Isis scrub), is now wholly under sugar-cane. It is a volcanic area, the red soils being derived from andesites, basalts, and tuffs which overlap the Mesozoic series towards the east and porphyries towards the west. This area extends westwards to Cordalba, and is wholly utilised for sugar growing. Passing along the main road S.W. from Childers towards Dallarnil, one soon goes out of red soil on to poor sandy loam with spotted gum, bloodwood, and grey gum, which overlies rhyolitic porphyries and granite porphyries, forming a vast igneous mass grading from volcanic to hypabyssal, and even to plutonic in places. These rocks extend unin- terruptedly from the western edge of the Childers red soil area to Sandy Creek (the Gregory River). The porphyry mass extends southwards in almost uninterrupted continuity to Biggenden Bluff, and I have seen rocks of similar nature crossing the Nanango line not far west of Theebine Junction. GEOLOGICAL NOTES — CHILDERS TO MUNDUBBERA. 141 The porphyries are in some places quite rhyolitic in texture; in others, whitish quartz porphyries ; in others a white, highly acid granite. At Biggenden Bluff they are granite grading into rhyolite. In the Woowoonga Range we have the same series extending from the Childers area to Biggenden Bluff in a north-south direction. The same intrusives burst through the Tiaro coal measures in places, accounting for the graphitisation of Mesozoic coals at Mount Bopple. They were intruded in post-Triassic but pre-Cretaceous times. Between Biggenden and Degilbo and between Degilbo and Didcot, as will be seen later, comagmatic intrusions take the form of granite dykes and bosses on the walls of which important mineralisation has taken place. West of Sandy Creek (Eureka Crossing) on the Dillarnil road, sedimentary rocks, such as slates, cherts, schists, and limestones occur. They are steeply inclined and in many places gold-bearing. The Paradise, Shamrock, Biggenden, Chowey, Stanton-Harcourt, and other minor goldfields are in this area. The Biggenden, Booyal, Marule, Stapleton, Gigoomgan, and possibly Goodnight Scrub limestones lie in this belt, which is Permo-Carboniferous in age. The diorites (altered to diabases) intruding the sedimentary rocks are late Palaeozoic and the granites are Mesozoic. Vast areas of diorite yielding beautiful soils occur on the Burnett at Booyal, and probably the same sequence obtains along the Burnett north to New Cannindah, where some of the finest soils are of dioritic or syenitic derivation. Degilbo to Didcot. — For several miles west of Degilbo one passes through schist and slate country with diorite and diabase dykes and later granite porphyry dykes, the sedimentaries being presumably Permo-Carboniferous or Upper Carboniferous. We then cross a belt of andesitic, dacitic, and rhyolitic lavas which are probably coeval with the Biggenden Bluff granite. They have flowed over Permo-Carboni- ferous strata and are in places gold-bearing, as at Mount Steadman, where hot-spring action has been at work in the igneous period. On the old Achilles and Gebangle goldfields, west of Mount Shamrock, gold was deposited by hot-spring action on the walls of porphyry dykes of this series, and the Mount Shamrock and Mount Ophir gold appears to have had a similar origin. There is much similarity between the gold and gold matrix on these fields and the Coolon and Kidston goldfields, as far as geological environment is concerned. Didcot to Gayndah. — West of Didcot we soon get on a range of very acid white granite yielding poor soils, which extend almost to Geroomba. The sandy soils carry a vegetation of stringybark, mess- R.S. N. 142 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. mate, sugar gum, spotted gum, red bloodwood, Moreton Bay ash, a narrow-leaved ironbark, and grey (blue) gum ( Eucalyptus rostrata ), contrasting markedly with the dacite flora of “dead finish” (Acacia sp.). In the bed of Deep Creek, near Degilbo, we have abundant pebbles and boulders of greenstone and chert, indicating the nature of the drainage area. In Gooroolba Creek we have boulders of metamorphosed basic rock, schists, and granite. The sedimentary rocks, which have a very old Devonian appearance, are intruded by granite as well as older diorite dykes. Pebbles in the creek are composed of amphibolite schist com- parable with Mount Mee. West of Gooroolba we have basic lavas (basalts) predominating on the surface. At Byrnestown we have basic igneous rocks and later rhyolites continuing as far as Wetheron. Between Wetheron and Barambah Creek we have extensive areas of red rhyolite porphyry alternating with metamorphosed basic rocks and schists. On the quartz porphyry the timbers are poplar box and silver-leaved ironbark • on the dacite, narrow-leaved ironbark and poplar box; and on the Tertiary (?) basalts, brigalow, bottle tree, and vine and turkey bush scrub. At Barambah Creek, near Gayndah, massive porphyries of a felsitic nature are seen on the road, and they are intruded by a later (probably Tertiary) basalt. As the town of Gayndah is approached we pass off the rhyolites and felsites on to a belt about two miles east of the town, consisting of conglomerate, sandstone, and shale of Mesozoic age, dipping south or south-west at about 20 deg. to 30 deg. These beds are not meta- morphosed. Right in the road a dyke of rhyolite porphyry is seen to cut these beds. It weathers spheroidally and is petrologically akin to the dominant porphyries and granites of the whole area so far described. This is primd facie evidence of the post-Triassic age of the Shamrock- Gayndah granites. West of Gayndah we pass on to porphyry and granite porphyry country with poplar box and silver-leaved ironbark. Here and there they grade into dacite, with narrow-leaf ironbark and poplar box. Occasionally basaltic intrusives occur with brigalow, bottle tree, silver- leaf ironbark, poplar box, and yellow bloodwood. The Binjour Plateau, between Gayndah and Mundubbera, and adjacent ranges consists of basic lavas and tuff, probably of Tertiary age. These volcanics, according to accounts given to me of bore strata GEOLOGICAL NOTES — CHILDERS TO MUNDUBBERA, 143 passed through, overlie in places metalliferous metamorphic rocks ; in other places coal-bearing (Mesozoic) sedimentaries. The lavas *are certainly post-Triassic and may be Tertiary. West of the Bin jour Plateau with its rich soils we have an area of highly metamorphic rocks, consisting of cherts, metamorphosed tuffs and andesites, crystalline limestones ranging from white to dark-blue in colour, intrusive diabases (altered gabbros), and later porphyries and granites. All except the last mentioned are highly folded and .strike N.E., and are probably Devonian, the same as the Mount Perry series. In the Mundubbera district we have granites of a red colour, some- what gneissic, belonging to an older series, and later porphyries belong- ing to the Mesozoic intrusive series. The latter was probably the one accounting for the sporadic gold occurrences. Granites and sedimentaries of older Palaeozoic age extend from Mundubbera to Pannes, to Hawkwood, to Cracow, to Auburn, and to Camboon. This vast area consists mainly of Devonian rocks and granite. West of Camboon, in the vicinity of Cracow and along the Dawson Valley, we get once more, as shown in my report to the Government, as yet unpublished, Lower Bowen beds steeply inclined but not highly metamorphosed, and dipping west. The Auburn-Rannes belt is a geanticline of great magnitude. Probably over 10,000 feet of sedimen- tary strata have been removed in post-Jurassic times.. At Auburn, Cockatoo, Nathan Gorge, and other places in the Taroom district, as I have shown in the unpublished report above mentioned, we have Ipswich beds, dipping in under the Walloon Creta- ceous series of the Darling Downs. In my paper on the East Moreton, in the Proc. Linn. Soe. N.S.W. 1906 (Pt. 3), I hinted that the Eumundi igneous rocks and the Noosa granite were post-Triassic and pre-Tertiary. Mr. Morton’s work has vastly extended the granitic intrusions of Triassic and post-Triassic age in Queensland. Conclusions. — I have come to the conclusion that west of Mary- borough we have an immense igneous complex, extending through to Taroom, which requires much future work to decipher thoroughly. But it is my opinion that while most of the basic intrusives in the belt extending from Biggenden to Gayndah are Palaeozoic, the rhyolites, dacites, quartz porphyries, and granites are in the main Mesozoic. 1 have also some to the conclusion that in the Mundubbera-Hawkwood district we have granites of two ages — viz., an older Devonian or Carboniferous granite and a later Mesozoic granite. West of Camboon the Mesozoic rocks are entirely unaffected by igneous intrusions. 144 PROCEEDINGS OF THE ROYAL SOCIETY OF QUEENSLAND. The various Mesozoic areas shown as Walloon on the present (Mineral Index) geological map of Queensland in the Burnett region should be designated Ipswich. It is the Ipswich rocks alone that approach the confines of the great Mesozoic basins in South Queensland. They are most likely to be preserved as inliers in the older series. Later geological work will, in my opinion, conclusively prove that all Mesozoic inliers between the Darling Downs and the Callide coal field consist of beds of Ipswich age. It is my conclusion, too, that in Southern Queensland a major geanticline of post-Triassic age (in fact post- Walloon) extends from the Dalveen district almost north-south under the eastern flank of the Darling Downs through Camboon and Rannes. A later subsidiary (and early Tertiary) anticline was developed east of this along the Mount Jubbera Range, between Boonah and Beaudesert south of Bris- bane, and along the D ’Aguilar Range north of Brisbane. These anticlines were the result of a thrust coming from the east, from the Pacific. Their formation was accompanied by faulting and volcanic manifestations — e.g., the post -Walloon diorite intrusions in the West Moreton (probably coeval with the older basalts), and the later mid-Tertiary trachytes and basalts. The Royal Society of Queensland, ABSTRACT OF PROCEEDINGS. Report of Council for 1923. To the Members of the Royal Society of Queensland. Tour Council has pleasure in submitting its Report for the year 1923. During the year twelve papers were published. With two exceptions, these papers were read and discussed before the Society. The following lectures, which were well attended and to which the public was invited, were delivered: — Mr. H. A. Longman, “ Australian Marsupials”; Dr. A. Jefferis Turner, “The Origin of the Australian Fauna”; Prof. Cossar Ewart, “The History of Feathers and the Breeding of King Penguins” and “The Scientific Breeding of Sheep”; Mr. B. Dunstan, “Some New Ideas on the Artesian System.” To the University of Queensland the Society is indebted for providing accommodation for meetings and for housing the library. As in previous years, the Government of Queensland voted £50 towards the Society’s activities, and we wish to acknowledge the practical assistance which was thus afforded. Appreciative acknowledgment is also made to the Walter and Eliza Hall Fund for a substantial subsidy towards the printing of a paper by Prof. T. H. Johnston and Mr. G. H. Hardy, entitled “Observations Regarding the Life Cycle of Certain Australian Blowflies. ’ ’ Owing to the increased cost of printing and the Society’s limited income, it was found necessary to curtail the publication of papers. In view of these circumstances a deputation, consisting of the President and Professors Richards and Goddard, interviewed the Premier (Hon. E. G. Theodore) with a request for further financial support. The deputation was sympathetically received, and the Premier generously granted its request to subsidise the Society ’s funds to the extent of £ for £ of the amount spent by the Society in printing, up to a maximum of £150 in any one year by the Government. The membership roll consists of 91 ordinary members, 9 life members, 8 corresponding members, and 5 associate members. During the year 21 new members were elected and 4 members resigned. R.S. — O. VI ABSTRACT OF PROCEEDINGS. There were 11 meetings of the Council. The attendance was as follows: — E. W. Bick 9, W. H. Bryan 10, W. D. Francis 9, E. J. Goddard 4, E. H. Gurney 8, H. A. Longman 7, E. O. Marks 11, H. J. Priestley 7, H. C. Richards 5, E. IT. Swain 4, R. A. Wearne 5, C. T. White 9. The desirability of increasing the membership of the Society is recognised by your Council. For this purpose a Special Committee, consisting of the President, Vice-Presidents, Hon. Editor, and Hon. Secretary, has been formed. This Committee has under consideration the distribution among suitable persons of a circular drawn up by the President, and emphasising the need for new members. The unsatisfactory condition of the Society’s library has been a. subject of consideration. It was decided to erect new shelving, but owing to the collapse of the ceiling immediately under the room in which the library is housed the proposed additional accommodation has been delayed. The Council notes with satisfaction the successful appeal by the Barrier Reef Committee for funds for research investigation. Up to date £1,300 has been collected in the form of private subscriptions. This amount, together with a Government subsidy, has enabled the Committee to appoint a full-time salaried officer. E. 0. MARKS, President. W. D. FRANCIS, Bon. Secretary. ABSTRACT OF PROCEEDINGS. Yll O z < J (0 z 111 111 D Q Li o > F W O 0 0) J < > o a iii x h ^ cocscooooot-oorH y! WW^OO(MIOOOIM J i— I rH rH rH rat (M l> 00 r- ILOOiHOO ^ CO O fc- J M * _ to ^ O O . ^ H ■g P P © « £ >> b & „ 2, § § bOOfcjo'Ho © , ^ W « - O .S | d H ^ ^ Q pi 2 -£ -JH rS g g .& 2 2 nl o CO ^ £ .5 ts s © S .S o ‘3 ^ bjD « -S © Pi 'M ^ o ® ^ t S ft \P S 3th 2 h Qj ffl r2 Hg , C3 P> o0 !> ,ogg>gSPiO© ift 0 oOOHS^cdccJXrO & o pKM^mPPPO P _© ’ft o O © ft c3 P • CO • CM OS © 173 P rt SPP sp cs m '£ W +3 P 3 ^ P m cS ® tPP 2 3 P "g rQ <1 CO M '•’•> fcJO .2 © pi H-> H cS 5 3rP Pi © o 0 ft X POP re co CO CO O O O 0 0 CO =c CO os 00 O O CO 0