CHARLES STEWART MIDDLEMISS  1859-1945 (Geological Survey of India 1883-1917)

 

Middlemiss graduated from Cambridge in 1880, a year after Tom La Touche, and 7 years before Philip Lake all of whom joined the Survey of India. Their Professor at Cambridge was Thomas McKenney Hughes.

 

The following is a 1945 obituary by L. L. FERMOR [Obituary Notices of the Royal Society, 14(5), 63-Nov 1945] with annotations and links to additional materials. All headings and comments in blue below are not in Fermor's account. Click here for Bibliography. 

 

        With the death of Charles Stewart Middlemiss on 11 June 1945, Indian geology has lost its doyen, a man of great versatility who lived an active life up to within some two months of his death in hospital in Tunbridge Wells. Middlemiss was the son of Robert Middlemiss of Hull and Elizabeth Findleyson Gale and was born in Hull on 22 November 1859. He was educated at Caistor and St. John's College, Cambridge, the college from which so many good geologists have come. He took his B.A. degree in 1881 and joined the Geological Survey of India as Assistant Superintendent on 21 September 1883, when H. B. Medlicott was still head (then known as Superintendent) of the Department. As Medlicott joined the Department in 1854, but three years after its foundation by Dr. Thomas Oldham in 1851, anyone who knew Middlemiss could in conversation go back in anecdote almost to the Department's foundation. With the death of Middlemiss the doyen of Indian geology is now Mr. Philip Lake who served for a short period from 1887 to 1891, followed closely by Sir Thomas Holland, who joined the Geological Survey of India in 1890.

The photo on the left was taken when Middlemiss joined the Geological Survey of India in 1882 (Courtesy of the Director Geological Survey of India), and the one on the right from c. 1930 is from a photograph provided by his daughter for Middlemiss's extended FRS obituary (written by Fermor).

 

       During his service Middlemiss did field work in the Himalaya, the Salt Range and Hazara; in Coimbatore, Salem and the Vizagapatam Hill Tracts in the Madras Presidency; in the Shan States and Karenni in Burma; in Bombay, Central India and Rajputana, and finally in Kashmir. During spells at Headquarters he was Curator of the Geological Museum in 1898-1899, and in charge of the Headquarters Office during 1907-1908. He was promoted Deputy Super­intendent in 1889, and Superintendent in 1895. He officiated as Director of the Department for two periods, in 1914-1915 and in 1916. He paid a visit on deputation to Ceylon in 1903.

        According to normal custom Middlemiss should have retired from the Geological Survey of India on reaching the age of 55 in 1914; but on account of the war and the difficulty of then obtaining recruits his service was extended, so that he did not leave the Department until 1 April 1917, after being decorated with the C.I.E. in 1916. As a result of this Middlemiss ranks second for long service in the Geological Survey of India, over thirty-three and a half years, the record being held by William King with a total service of over thirty-seven years (due to his retention in the Department until the age of sixty). Middlemiss then went direct from the Geological Survey of India to the service of His Highness the Maharaja of Kashmir and Jammu , with the title of Superintendent, Mineral Survey of Jammu and Kashmir State. He did not retire from Kashmir service until his seventy-first year, when, in the recess period of 1930, and after a total service in India of over forty-six years, he gave much pleasure to those geologists who had not met him by returning to Calcutta to collect a few personal possessions and say farewell to the focus of so much active life.

   

HONOURS Of other scientific activities we may record that Middlemiss became an ordinary and life member of the Asiatic Society of Bengal in 1884, and a Fellow in 1912. At the time of his death he was by four years the senior member of the Society (now with the appellation 'Royal'), and also the senior Fellow.  He became a Fellow of the Geological Society in 1900 and was awarded the Lyell Medal in 1914. He was elected a Fellow of the Royal Society in 1921. He was President of the Geological Section of the Fourth Indian Science Bangalore in 1917. He was President of the Ninth Indian Science Madras in 1922, and of the Mining and Geological Institute of India in 1928. He became a Fellow of the Royal Society of Arts in 1934 and of the American Geographical Society of New York in 1936.

 

MARRIED LIFE AND CHILDREN In 1886 Middlemiss married Martha Frances Wheeler, daughter of the late General F. Wheeler, Bengal Staff Corps. They had five sons and one daughter. Two sons died young in India, two were killed in the last war, one in Gallipoli and one in France, and the fifth, Hugh Percival, is pursuing a musical career. On retiring from India Middlemiss settled down at Crowborough Sussex, where he became a centre of local activities. Mrs Middlemiss died in July 1931. To his daughter, Mrs. Forward, who has kept his house for him during this war, I am indebted for help in preparing this notice, and for his photograph.

    

CHARACTER Before discussing Middlemiss's work it will be pleasant to give some account of his personal characteristics. As I and my contemporaries first knew him was already in the forties-partially bald with an aureole of grey hair, bright blue eyes, pleasant friendly expression, very sturdily built and above medium height. He had an extraordinarily youthful mind, was always ready for fun, and was often the life and soul of parties, especially if youngsters were present. Women liked him for his courtesy and chivalry.

       Middlemiss  had an experimental mind and was willing to try anything and especially anything new. The result was that he had a multitude of interests outside geology, the principal of which were sketching and music, as recorded in Who's Who. But he was also interested in verse, in Esperanto, in chess, and in rowing and in lawn tennis, which he was playing in local tournaments until he was eighty.

 

ChorART Of these interests the most important was sketching in water colour. From India he recorded in this medium the incidents of camp life and travel and the scenery amidst which he worked. Some of these sketches are very amusing and some very charming. Fortunately he was induced recently to arrange these skethes and to list of these sketches chronologically in a portfolio, which is now in the possesion of his daughter. Amongst his effects there were also some framed sketches  and one of them will, it is hoped, be reproduced in a forthcoming issues ­of Notes and Records [This appeared in Notes and Records of the Royal Society of London, Vol. 4, No. 2. (Oct., 1946), pp. 214-215]. It is a view of the Chor Mountain, clearly visible from Simla, and done in 1884. According to the old standing orders of the Geological Survey of India, officers of the Department were required to illustrate their reports with drawings and sketches, and only if they were so incompetent as to be unable to do this were they, as a concession, allowed to use photography! As a result, in my day, all officers of the Department were provided with a camera lucida as a part of their camp equipment to enable them to sketch accurately the outlines of hills and mountain scenes. Several of the earlier officers used this aid with great effect, e.g.Thomas Oldham, Griesbach, and,  facile princeps, Middlemiss.

      A magnificent example of what can be done by a man of combined artistic and scientific temperament is Middlemiss's own memoir on the Geology of Hazara and the Black Mountain (Mem. Geol. Surv. India, 26, 1896). Amongst the numerous illustrations are some panoramic views of which the outlines were drawn with the aid of a camera lucida; to the panorama the geological formations have been added using the same colours as in the geological map illustrating the memoir. Plate 9, to choose one of several, is a superb example of the way in which geological reports can be illustrated so as to give results far more effective and explanatory of the scenery than are obtainable by photo­graphy. Apart from the multitude of the usual geological sections and other illustrations that this memoir contains, all drawn of course by the author, we may mention specially two plates of micro-sections of rocks hand-drawn by Middlemiss and much clearer in elucidating the composition and structure of the rocks depicted than the majority of micro-photo graphs that are published. In this last respect Middlemiss agreed with Harker, who also preferred to draw his own micro-pictures of rock sections.

 

MUSIC Concerning Middlemiss's second hobby, music, we must note that he comes of a musical family. So it is not surprising that with his experimental mind, he should have looked at music experimentally. Not being myself a musician it is suitable that I should quote here a passage from a letter from Sir Edwin Pascoe that explains clearly this experimental outlook:  He thought he was fond of music and so he was in a way. But it was not so much music itself which attracted him but the various instruments which produced it, the method of scoring it, and any stage device for illustrating it. If Middlemiss were present at a performance of Wagner's opera Die Walküre. I can imagine him consumed with curiosity as to how the Valkyries on their horses were made to gallop over the clouds. While all true lovers of drama and music would be doing their best to forget that the figures in the stage sky were just dummies mechanically propelled with ropes and pulleys, and trying to imagine that they were real horses and horsewomen carrying the dead bodies of heroes to Valhalla. Middlemiss would probably been endeavouring to invent some more realistic method of depicting the scene.

        It is not surprising that Middlemiss invented a new musical notation (see Bibliography, 1935). The fundamental reason for this appears to have been his objection to the fact that in the usual notation the same note may be indicated in more than one way: e.g. that C sharp and D flat are rendered by striking the same key. On this notation Pascoe writes to me: His musical notation was not without merit, and he took it quite seriously. I think it failed chiefly in that you had to read and identify every note. In our ordinary system this is not necessary. In a rapid rise up or down the scale, for example, you do not read every note-you have not time-but are conscious of a regular sequence. What you have to look out for, in fact, is any irregularity or interruption in the sequence.

 

VERSE & ESPERANTO Middlemiss also amused himself with verse, and especially by experi­menting with various forms of verse. Whilst his results could not be described as poetry they were very tolerable and readable verse. He liked especially lyrical ingenuities such as rondeaux, rondels, ballades, triolets, etc. Middlemiss in his enthusiasm for new methods was an active exponent of Esperanto in its early days and acted as Honorary Secretary of the Calcutta Esperanto Society (or was it the Esperanto Society of India?) for some years. He tried to infect the Geological Survey of India with this new language and several of us allowed ourselves to be dragged at his chariot wheels. For some years, in the days when we had our lunch sent to office, Middlemiss, Vredenburg and I used to lunch together, and invariably our greetings and conversation opened in Esperanto before reverting to English.  On my first period of leave, in 1908, I visited Geneva, immediately after an International Esperanto Congress had been held there. I tried in vain to make use of the new language, and was forced to use my inferior French. On my return to India I reported the results to Middlemiss and Vredenburg and cut myself adrift from Esperanto. But Middlemiss carried on for some years. I have heard of one example of Esperanto proving useful. Hayden (later Sir Henry Hayden) when on duty in Afghanistan used it in writing to headquarters in Calcutta, for his letters if opened en route would be understood by no one!

     The difficulty that faces inventors of new languages like Esperanto, or even Basic English, or new systems of musical notation like Middlemiss's, is that even if that new language or system is better than the old, it stands no chance of replacing existing languages or systems. For by learning the new you cannot thereby avoid also learning the old, for the simple reasons that all preceding work has been published in existing languages or notation, and that there is not the remotest possibility that the whole corpus of these past works will be translated into the new tongue or rewritten in the new notation.

     To young officers in the Department Middlemiss was very helpful, both in the field and at Headquarters. I did not have the advantage of working under him in the field, but I must recall his kindness to me when I offered my first paper for publication by the Geological Survey. Sir Thomas Holland, then Director, had introduced the custom of circulating papers for criticism amongst officers, starting with the most junior and ending with the most senior, before it went to the Director himself. My paper reached Middlemiss near the end of its round, and in it he, being a master of English composition, found many passages that could be better expressed. Instead of commenting on this on the file he brought the manuscript to me, and we went through it together and made the necessary improvements before the paper went further.

       In spite of his obviously happy nature Middlemiss occasionally had violent fits of temper, a fault of which no one was more conscious than he. He was never above telling stories against himself, and the difficulties his temper got him into. One story may be permitted here, as it illustrates Middlemiss. I am indebted for it to Sir Edwin Pascoe to whom it was told by Middlemiss himself. It is the story of Middlemiss's old bearer and the scones. One day in camp the old man angered him in some way, with the result that Mlddlemiss sacked him on the spot. The old servant duly packed up and disappeared. Now the old man was very clever at making scones and Middlemiss was very fond of scones for tea. That afternoon, however, there were no scones when Middlemiss returned from his work. When the next day came and there were again no scones Middlemiss began to wonder whether he had not been a little hasty. On the third day, to his surprise and delight, there appeared a dish of his usual scones, and soon after the old bearer himself carrying on as if nothing had happened. When asked how he dared to come back after he had been sacked, the old man's answer was; 'Well, Sahib, you and I have been Master and servant together for over twenty years, and I guess we shall go on being so until one of us dies!' And they did; it was the old man who died.

 

    

GEOLOGICAL WORK During his long term of service in the Geological Survey of India, Middlemiss spent but a small proportion of his time at Headquarters, except for the normal recess season. He was at Headquarters for a short time as Curator of the Museum and Laboratory and on another occasion was in charge of the offices; whilst at the end of his service he officiated twice as Director of the Department. Fortunately for the scientific world he was never a permanently Director, as his talents did not extend to administrative activty and official correspondence. Instead he was essentially a field geologist who liked a direct appeal to the facts of Nature as displayed in the field and elucidated by the hammer; though he was glad, of course, to confirm and support these results by work at Headquarters during the recess, by means of the microscope and specific gravity determinations. 

       His total service in India can be divided into three spells. The first, of some ten years' duration, was spent in North-Western India in the Himalaya, the Salt Range and Hazara. The second spell was twent-two years long and was devoted mainly to the problems of Peninsular India, starting with some ten years in the Madras Presidency (the Salem and Coimbatore districts and the Vizagapatam Hill Tracts) with a Burma and a Ceylon interlude, followed by some twelve years in charge of the Central India and Rajputana Party (later the Bombay, Central India and Rajputana Party) of the Geological Survey of India, with interludes for the headquarters duties mentioned above. From 1908 to 1913 Middlemiss combined winter work in Central India and Rajputana with summer work in Kashmir, so that on retiring from the Geological Survey of India in 1917 he was in trim for his new appointment as Superintendent of the Mineral Survey of Jammu and Kasmir, a post that he held until 1930, giving him a total third spell, in Kashmir, spread at intervals over some twenty-two years; the second and third spells thus overlapped, and the whole gave him a total of forty-six years' service in India. 

       The first and second spells were punctuated by earthquake investigations, namely the Bengal earthquake of 1885, the Kangra earthquake of 1905, and the two Calcutta earthquakes of 1906.  Middlemiss's most fruitful work was done in North-Western India, scientifically during his first spell in the Himalaya and Hazara, and scientifically and economically during his third spell in Kashmir, whilst his investigation of the Kangra earthquake during his second spell acted as a mountain refresher. The middle spell was not so fruitful, at least as represented by published work, and Middlemiss appears to have been less happy upon the Archaean rocks of the Peninsula, where petrographical studies play such a large part and where it is difficult to work successfully on a broad scale, than he was in Extra­Peninsular tracts, mainly occupied by younger formations, where broadr stratigraphical problems prevail and where the age and relationships of the rocks in most cases have not been so greatly obscured by the operation of intense metamorphic processes.

 

JANUARY 1884 WITH OLDHAM TO JAUNSAR For his introduction to Indian geology Middlemiss was sent out with R. D. Oldham, whom he joined in the field in January 1884. Oldham was working in the Jaunsar Bawar tract of the Lower Himalaya behind Dehra Dun, and as a result of this experience Middlemiss saw not only the Himalayan rock formations of Oldham's tract, but also the sub-Himalayan tract of Dehra Dun described by H. B. Medlicott·in his classic memoir of 1864 (Mem. Geol. Surv. India, 3). Whether Middlemiss actually got as far west as the Chor Mountain in the Simla Himalaya does not appear, but that he went close thereto is proved by the water-colour sketch referred to on page 264, the sketch being dated 1884; and by the reference in his third paper (Rec. Geol. Surv. India, 22, 29). From Chakrata in Jaunsar Bawar Middlemiss marched through Tehri Garhwal to British Garhwal to the east, which with Kumaon still farther east was to form his hunting ground for some six field seasons.

      Middlemiss's work in Garhwal and Kumaon may be divided into two sections, namely that on the Himalayan formations, mainly pre-Tertiary, constituting the Lesser Himalaya, and that on the Sub-Himalayan formations, mainly Siwaliks, forming the outer fringe of the Himalaya. The latter work was done the later; but as it was brought to complete fruition in Middlemiss's important and illuminating memoir on the 'Physical Geology of the Sub­ Himalaya of Garhwal and Kumaon' (Mem. Geol. Surv. India, 24, Part 2), a memoir that has already assumed the position of a classic illustrated with classic sections, it may be noted first. In this work he mapped completely the Sub-Himalayan tract from the Ganges on the north-west to the Sarda river on the Nepal frontier to the south-east,his survey being designed as an extension in a south-easterly direction of the geological work described by Medlicott. Owing to the possession of better maps-partly on the scale of l-inch to the mile and partly, in the Reserved Forests, on the scale of 4 inches to the mile­ Middlemiss was able to work in much greater detail than Medlicott, so that he was able, using his own modest phraseology, to amplify and modify Medlicott's work.

 

STRUCTURE OF THE HIMALAYA The essence of Middlemiss's work was to show that the Sub-Himalaya is divisible into roughly parallel strips of Siwalik strata separated by reversed faults so that in every case a younger division is dipping under an older one to the north-east. At one cross-section he maps five successive reversed faults, of which two are within the Siwaliks, one separates the Lower Siwaliks (Nahans) from the Great Limestone of the Lesser Himalaya to the north, and the other two, still farther north, separate zones of the Himalayan formations. He regards each of these reversed faults as representing approximately the position of an ancient shore-line or mountain-foot so that these reversed faults are at the same time in a certain measure limits of deposition for the formation immediately south of each (loc. cit. pp. 118-119). They are also successively younger as one approaches the plains, corresponding with the deposition of later sediments of the plains side of the range. Consequently the range continues to grow by fresh additions to itself, like a coral, by the incorporation of its offspring with itself (loc. cit. p. 133). His reasoning leads him also to suppose that there must be a reversed fault between the outermost belt of Siwaliks and the Gangetic alluvium and that consequently the Himalayan range is still growing. Further, he is confident that before the deposition of the Siwaliks began the Himalayan range was already in existence much as we know it now, and that it is highly probable that a barrier of crystalline rocks existed in Tertiary times between the Sub-Himalayan deposits of this side and those of Hundes (loc. cit. p. 115).

 

OSMOND FISHER AND ISOSTACY As illustrating Middlemiss's  readiness to use new ideas,  it is interesting to note that in discussing the bearing of his work on the theory of mountain formation he welcomed Osmond Fisher's Physics of the Earth's Crust, first published in 1881, in which is proposed the hypothesis that the crust of the earth is in a state of approximate hydrostatical equilibrium so that the Himalayan mountain range has been rising pari passu with decrease of load, as by denudation it supplied sediment to the plains at the foot, which them­selves have continued to sink under the burden of increasing load. The term isostasy1 now used to describe this state of hydrostatic equilibrium was not proposed by Dutton until 1889, and this term had evidently not reached Middlemiss when he wrote his memoir, published in 1890.  Both Osmond Fisher and Middlemiss, however, must be described as isostasists, and in later works Middlemiss uses the term freely.

 

Pascoe's Footnote #1.  It evidently took some time for this term to reach India. For R. D. Oldham*, influenced perhaps by Middlemiss's writings, introduced Fisher's views in discussing the uplift of the Himalaya in the second edition of the Manual of the Geology of India, published in 1894, without using this term. Strange to say, this term has not yet reached the Shorter Oxford English Dictionary (in two volumes  second edition 1933, reprint of 1939). It does, however, appear in adjectival form in Chamber's Twentieth Century Dictionary, dated 1903.

*comment: Oldham's first mention of any of Fisher's theories is in an article published in 1884 in the Asiatic Society on fossil synchroneity, in which he discusses conflicting evidence for changes in latitude through paleoogeographic climate change. Middlemiss joined Oldham in the field in January 1884 and it is very likely they discussed Fisher's theories at this time.   Fisher had published a Geological Magazine article on changes of latitude as early as July 1878 which Oldham might have read before leaving for India.  The first edition of Fisher's book was published in 1881, and its second enlarged edition in 1894.

     

 

ANCIENT RIVERS AND GRAVELS In our obituary notice of Dr. G. E. Pilgrim we mentioned the suggestion advanced independently by Pilgrim and Sir Edwin Pascoe, and based on two quite different lines of approach, that the three main rivers of northern India, the Brahmaputra, the Ganges and the Indus, were once part of a single river rising in Assam, flowing north-west along the foot of the rising Himalaya and then turning south-west towards the Arabian Sea, the course being demarked demarcated by the distribution of the Siwalik Boulder Conglomerates. This river was called the Siwalik river by Pilgrim and the Indobrahm by Pascoe.  Middlemiss's memoir contains a detailed survey of a portion of the course of this supposed river. I do not know if Middlemiss accepted this idea of Pilgrim and Pascoe, but it seems necessary to note here that his memoir contains a passage that appears to disagree with their hypothesis. He writes (p. 120):

 

Nothing is more clearly demonstrated in the whole range of Sub-Himalayan geology than the connexion between the position of the debouchure of the present large rivers and the deposits of Siwalik conglomerate. That being so, we are bound to believe that these deposits were formed by the direct parents of those rivers, in the places where they are now found.

 

This means that the separate patches of Siwalik conglomerate shown on Middlemiss's map cannot be regarded as sections of a continuous strip of conglomerate parallel to the Himalaya, but coincide in position with the places where the ancestors of the present Ganges and Kosi emerged from the hills2. {Footnote 2 Pilgrim, working on the Panjab section of the Himalaya came to a different conclusion and found the conglomerates continuous between the debouchures of the transverse Himalayan streams}.

 

NAPPES Before he made his studies of the Sub-Himalayan belt, so happily brought to fruition in the memoir discussed above, Middlemiss had begun his survey of the belt of Lesser Himalayan formations lying to the north of the Sub­Himalayan tract, work which, though continued at the appropriate season of the year for as long as he was engaged in this part of India, was not carried to completion before he was detached for work elsewhere. Nevertheless, Middlemiss's work on the Himalayan formations is discussed by him in a series of papers in the Records of the Geological Survey of India, of which the most important is his 'Physical Geology of West British Garhwal' (Rec. Geol. Surv. India, 20, 26-40), published in 1887. This paper contains the results of a very careful survey on the 1-inch scale, with his published map on the quarter-inch scale, of a tract of country the geology of which was previously unknown. This paper records in fact the results of the first detailed geological survey of any portion of the Himalaya. The tract described contains an elliptical area (within which is Kalogarhi mountain composed of gneissose granite), arranged parallel to the Himalayan axis, of schistose rocks (Middlemiss's Inner formation) encircled by a belt of unmetamorphosed sedimentary strata (Middlemiss's Outer formations) shown by Middlemiss to be at least in part of Eocene (Nummulitic) and Mesozoic age (Tal beds). Middlemiss decides that the Inner schistose rocks are older than the Outer, unmetamorphosed, formations, in spite of the fact that his survey disclosed a very remarkable arrangement of the rocks, namely that everywhere round the ellipse the encircling younger Outer formations dipped towards and apparently under the Inner formation, the older schistose series, so that it was (p. 36):

 

difficult to get rid of the first impression already alluded to that the whole is asynclinal trough with the Outer formations below, and the Inner above. One seems almost driven to conclude that if a boring were sunk through the centre of the schistose area, we should inevitably strike the Tal beds below.

 

Middlemiss resisted this conclusion and instead explained the structure by showing that everywhere the younger outer formations dipped under the older inner formation along reversed faults or thrust planes: in fact he appears to have regarded the Outer formations as being everywhere tucked in under the Inner formation without supposing that this reversed relationship extended continuously under the syncline.

 

INVERTED METAMORPHISM In our obituary notice of Pilgrim we noted how that great geologist Medlicott in his discussion of the stratigraphy of the Simla Hills was 'at first tempted to look for grand inversion of strata', and how by failing to yield to temptation he missed the truth and left the problem for Pilgrim and West to solve some sixty years later. If the second great Himalayan geologist, Middlemiss, had accepted the conclusion that his hypothetical boring would have struck the younger rocks under the older in the middle of the syncline he would not have left the problem to be solved by J. B. Auden (Rec. Geol. Surv. India, 71, 407-433 (1937)) some fifty years later. The combined work of Pilgrim, West and Auden has shown us that the anomalous relations of geological formations on the grand scale in the Himalaya are to be explained in terms of repeated overthrust nappes of formations thrust bodily outwards, i.e. south­westerly, from the Central Himalayan direction for many miles over the autochthonous formations of the Outer or Lesser Himalaya, with subsequent denudation producing scattered windows through which scattered views of the  underlying formations can be obtained. How many of us, if working as long ago as Medlicott and Middlemiss, would have stood on the brink of these great discoveries and not have missed them!  But these two geologists laid solid founda­tions on which others have built. Auden records the excitement with which he and West independently read Middlemiss's paper on West Garhwal and wondered if he was really describing a nappe without realizing it.

        In this West Garhwal paper, and a second paper (Rec. Geol. Surv. India, 20, 134-143) discussing the geology of the Dudatoli Mountain, Middlemiss describes the distribution of a garnetiferous aureole round the intrusive gneissose granite and recognizes the strong resemblance of the Kalo garhi (Lansdowne), Dudatoli and Chor mountai ns both in rock composition and in geological structure.  He regards the granite as intrusive in the schists, but is very cautious on the source from which it was derived. We know now that these granites with their associated schistose rocks form parts of nappes of strata derived from the north-east.

      From Middlemiss's Memoir on the Geology of the Sub-Himalaya one learns that he hoped later to write a comprehensive memoir on the geology of the Himalayan belt of Garhwal and Kumaon, of which his papers in the Records, including the two just discussed, were forerunners. His transference to the Salt Range and Hazara before he had completed his field surveys of Garhwal and Kumaon would not have been a geological tragedy if he had been allowed later to return to the Himalaya and complete the work for this memoir.  Had he done so, would he have failed finally to arrive at the truth and recognize the large-scale overthrusting, instead of leaving the problem unsettled for fifty years?  Truly the responsibilities of Directors of Geological Surveys are some­times very great, especially when they have to balance the rival claims of a short-term policy directed to the immediate appraisal of the value of mineral deposits of possible economic value against those of a long-term policy directed to the understanding of the geology of a country, upon which ultimately depends knowledge of the distribution of the known mineral deposits and of the possibilities of additional discoveries. (See Rec. Geol. Surv. India, 24, 27.)

 

MOBILITY OF SALT     The discovery by Warth, in 1888, of Cambrian trilobites in the Salt Range of the Panjab was the cause of Middlemiss being sent there in 1889 and 1890 to investigate this and other problems that had arisen in that field of research. As a result many more trilobites were found and Middlemiss produced a brilliant paper on the geology of the Salt Range 'with a re-considered theory of the Origin and Age of the Salt Marl' (Rec. Geol. Surv. India, 24, 19-42 (1891)). At the time of his visit Wynne's view held the field, namely, that the Saline series, a term used to designate the Salt Marl and its associated gypsum and rock salt, was in its normal stratigraphical position at the foot of the Salt Range, beneath the Purple Sandstone, and therefore either Cambrian or pre­Cambrian in age. From Middlemiss's careful sections and descriptions it is seen, however, that the Salt Marl never shows a normal sedimentary contact with the overlying Purple Sandstone nor with overlying younger beds (Car­boniferous or Tertiary). Having devoted a large portion of his paper to exact description, in his usual methodical manner, Middlemiss th en allows his speculative faculties full play and considers the implications of the 'anomalous quasi-intrusive relations' of the Salt Marl to other formations. He asks (loc. cit. p. 42):

 

     Can we see in it anything of the nature of a scum, such as we might picture to ourselves as having partly secreted at the surface of an ancient untapped magma, and partly resulted from that secretion by induced changes in the overlying dolomitic strata?

       If we can, we have but to give the substance a gently intrusive or injective impetus, followed by consolidation, some time during the Tertiary period, to account for all the otherwise perplexing circumstances under which the salt-bearing beds of the Panjab are found.

 

   

From Middlemiss's paper it will be seen that the date of the disturbances that he interprets in this manner must have been post-Middle Siwalik (post­ Miocene) .

Since Middlemiss wrote, intrusive salt domes and plugs have become a commonplace of geology and such domes, composed of salt of Cambrian age, but of late tectonic production, have been found widely distributed in Persia; it is now considered unnecessary to postulate the existence of a subterranean magma, and geologists are satisfied that the incompetence of saline beds under tectonic pressure provides an adequate explanation of the phenomena so graphically described by Middlemiss.

 

SALT RANGE SALT AGE CONTROVERSY    With this paper Middlemiss initiated the controversy that is still unsettled concerning the age of the Saline series of the Salt Range, whether it is Cambrian (or pre-Cambrian), or whether it is post-nummulitic in age and has acted as the sole of a large-scale overthrust of the whole mass of the Salt Range over Tertiary saline beds. According to the recent careful re-survey of the whole of the Salt Range carried out over several years by Mr E. R. Gee of the Geological Survey of India, the Saline series is of Cambrian (or pre-Cambrian) age and the abnormal relations to other formations so carefully described by Middlemiss are due to the incompetence and plastic nature of the materials of the Saline series; all the other stratigraphers who have examined the ground recently are in accord with Mr Gee on this interpretation. But in apparent support of the Tertiary age of the salt, Professor Birbal Sahni and his associates have discovered plant and other organic remains of Tertiary to Recent age apparently embedded in the Salt Marl and associated beds. The problem is still, therefore, subjudice, and to help solve it Professor Sahni has arranged two symposia, one of which has already been held and its results published. To the first symposium Middlemiss in his eighty-fourth year has contributed  a short note (Proc. Nat. Acad. Sci. India, 14, Sect. B, 267-268) in which he claims that his theory of :

 

intrusive Salt Marl of the Salt Range is in harmony with up-to-date requirements and need no longer be silently dismissed as a rash speculation; even though acceptance of the possible guess at the end of my paper that an igneous magma may have been ulti­mately responsible; is too much to expect.

 

The flat contradiction between the evidence of stratigraphy and that of palaeontology referred to above may eventually be resolved by the recognition that in view of the plastic nature of the saline beds under pressure, and the solubility of the salt, any fossil remains found therein may be exotic, and that Saline beds of Cambrian or pre-Cambrian age have been caused to behave in post-Miocene times in the quasi-intrusive manner described by Middlemiss and have thereby acquired Tertiary and Recent characteristics in the form of enclosed Ranikot foraminifera and plant remains of Tertiary and Recent age.

[The controversy is over: it is now generally accepted that Gee is correct and that Cambrian salt has incorporated later fossils through the process of flow. The salts permits the displacement of the Potwar plateau SSE at about 7 mm/yr]

 

Kashmir and the Hazara Syntaxis  From the Salt Range, Middlemiss, in 1890, was sent to Hazara, now in the North-West Frontier Province [of Pakistan], to study the coal of the Dore River, and this led to his being sent to make a continuous survey of the whole of the southern half of Hazara, following the previous disconnected and unfinished work of Wynne and Waagen, and including a visit to the Black Mountain country during a military campaign. This work lasted until 1893 and resulted in Middlemiss's splendid memoir on 'The Geology of Hazara and the Black Mountain', published in 1896 (Mem. Geol. Surv. India, 26).

    Although the mountains of Hazara belong to the Hindu Kush system of folding with its N.E. strike and are not geographically a part of the Himalayan system of ranges with their N.W. strike in Kashmir next door to Hazara, yet geologically the Hazara mountains are composed of rock formations con­tinuous with those of the Himalayan belts in Kashmir, the sharp bend of strike taking place north of the hairpin bend of the Jhelum River near Muzaffarabad.  As a result, in his survey of Hazara, Middlemiss was studying stratigraphical belts analogous to those already partially surveyed in Garhwal and Kumaon. As in Garhwal and Kumaon he divides his country into zones of disturbance or elongated blocks of formations that lie parallel to the general strike of the country. He recognizes four such zones (lo cit.p.87):

 

NNW  (A) Crystalline and metamorphic zone.

(B) Slate, or Abbottabad zone .

(C) Nummulitic zone.

SSE     (D) Upper Tertiary zone .

 

He calls A the 'innermost zone' and D the 'outermost zone', and describes and discusses each separately; and then, comparing them in respect of the amount of elevation, compression and denudation they have suffered, he finds that (loc. cit. p. 265):

 

 zone A has been most elevated, most compressed, and most denuded; whilst we travel south-east through the other zones in order we find that they have successively been less elevated , less compressed, and less denuded.

 

On the south side of each of the three northern zones of disturbance is a boundary fault by which each zone has been thrust over the next zone to the south. Middlemiss ranges the zones in order of age A > B > C > D and regards the boundary faults as similarly successive in age. Further, discussing how this succession has been produced, he writes (loc. cit. p. 280):

 

At stated periods as we have seen, after a certain packing of the  rocks had taken place, a great block of such rocks has yielded as a whole and gone sliding over the one to the south, or under the zone to the north, producing a thrust-plane, and marking off two disturbance zones from each other.

 

Middlemiss does not use the term nappe, but as he recognizes the similarity of the structural arrangements of Hazara to those of the Himalaya in Garhwal and Kumaon it seems probable that he is really describing, in the passage quoted above, overthrust nappes, and that a re-examination of Hazara by one familiar with nappe structure and modern terminology would show this. Also it seems to follow that had Middlemiss returned to his earlier hunting-grounds in Garhwal and Kumaon he would have revised his interpretation of such structures as the Kalogarhi (or Lansdowne) synclinal. [Note: Both nappe structures, if unfolded, and Middlemiss's sliding, require for their explanation that the original thrust-planes must have been of low angle. In a much later paper (1919) resulting from his surveys in Jammu and Kashmir , Middlemiss. succeeds In  measuring the inclination of the thrust-plane or reversed fault between the Siwalik and Miocene zones, of formations ncar Kotli in Jammu . He shows it to be from 12° to 15° only. (See Rec. Geol. Surv. India, 50, 122. He was the first to demonstrate this fact, which is all the more important because it relates to what is often termed the Main Boundary Fault of the Himalaya. (See also Fermor, Rec. Geal. Surv. India, 62,410 (1930).]

       In his discussion of his oldest zone in Hazara, the crystalline and metamorphic zone A, the only zone in which igneous intrusives have played a part, a zone in which the youngest formation is his Infra-Trias (now regarded as Permo-Carboniferous in age) and its metamorphosed equivalents (the 'Tanols or Tanawals'), Middle­miss points out that this is the only zone in which igneous intrusion has played a part, and that, therefore, the intrusion of the gneissose granite must have been pre-Triassic in date and post-Infra-Trias. Middlemiss's next zone, B, contains rocks ranging in age from his oldest series, his Slate series (Attock slates), to Tertiary, and none of them are intruded by the gneissose granite .

      Whether Middlemiss would have adhered to his view that the gneissose granite of Hazara was pre-Triassic in age, if he had considered that the only one of his zones of disturbance that contains intrusive gneissose granite was a nappe brought into position possibly from far to the north-west, is of course unknown. What he does do is to enlarge on the unity of the gneissose granite of Hazara with that of the Central Himalaya and thereby to ascribe a similar age-to the latter. This leads him to disagree with the views of MacMahon and others that the Himalayan ranges must be entirely Tertiary in age. As always, he is cautious, and not dogmatic in dealing with such a problem. Thus, whilst as before he urges (loc. cit. p. 284) 'the probable great age of the Himalaya as opposed to the popular ideas that they were the product of yesterday, geologically speaking', he guards himself against misconception. by referring to the far-reaching views of the Rev. Osmond Fisher, and says (loc. cit. p. 285) that it has been gradually becoming evident to all who really examine the question that the Himalaya are and have been in a constant state of change so that 'one might say literally that the Himalaya of to-day are not the same as those of yesterday'. But to be more precise Middlemiss writes :

 

Hence in speaking of the Himalaya of a past geological age or epoch we mean, or at least I mean, that old representative of them which held about the same position and acted functionally in the same way as does the mountain-range going by the name of Himalaya to-day. It may not always have been of the same height as the Himalaya of to-day . It may sometimes have been represented by long parallel coast lines, or by archipelagos with chains of mountainous islands following similar parallel li es, but that it kept certain original features, and that a mountain core recognizable in its unity" persisted t rough Tertiary, Secondary and possibly into Palaeozoic times, I have no doubt.

 

The above is summarized in the final sentence of this magnificent memoir:

Rome was not built in a day, neither were the Himalaya either.

As indicating the philosophic attitude of Middlemiss to every type of problem, it is worth mentioning that although the gneissose granite of Hazara has demonstrably acted as an intrusive rock to the sedimentary formations yet Middlemiss is prepared for this rock to be either a gneissose granite or a granitic gneiss (loc. cit. p. 278) and therefore would not regard the views of the French school now adopted by many modern investigators in Britain and elsewhere, namely, that some granites at least may be thoroughly metamorphosed and melted sediments, as incompatible with the field evidence. For he refers to the crystalline core of the Himalaya with the following words (loc. cit. p. 278): whether we consider it as a thorough granite or as a slumbering gneiss that at one time became functional as a granite. Middlemiss also draws attention to the uniformity of petrological characters of this granitic or gneissic rock right through the Himalaya from end to end 'as far as observations have gone', as well as in Hazara, and contrasts this fact with the state of affairs in Southern India, where 'there is no uniformity what­ever as regards the gneissic foundations of the peninsula' (loc.cit.p.275).

       On completing this work in Hazara Middlemiss was not sent back to Garhwal and Kumaon, as might have been hoped, but to the Madras Presidency, thus commencing his second or Peninsular period of work. Instead of next discussing this it seems better to refer first to his earthquake work and then to his third period in Jammu and Kashmir.

EARTHQUAKES       Northern India, as is well known, is situated on one of the earth's main earth­quake belts, the Indian section striking first north-north-east along the Baluchistan and North-West Frontier ranges and then south-east and east parallel to the Himalayan ranges as far as Assam, where it turns south-west to south through the Assam-Burma ranges. This is a zone of instability, residual from the Tertiary period, of over-thrusting of Extra-Peninsular India against the massif of the Peninsula (or vice versa), a period that still continues and will continue as long as the Himalaya and geologically related ranges are still in active movement and process of formation. In consequence, India is subject to earth­quake shocks, often of disastrous magnitude, at unequal and at present unpre­dictable intervals. In India the field investigation and study of earthquakes is one of the duties of the Geological Survey. Consequently, as it is imperative that the evidence presented by earthquake shocks should be recorded at the earliest possible moment, while it is still fresh, officers of the Geological Survey of India are liable to find themselves ordered by telegram at a moment's notice to the scene of some important and often disastrous earthquake.

         Middlemiss during his career had to investigate the violent Bengal earth­quake of 14 July 1885, two mild Calcutta earthquakes of 1906, and the disastrous Kangra earthquake of 1905. We need discuss only the Kangra earthquake, which was one of first magnitude and, as judged by loss of human life (over 20,000), one of the most disastrous of modern times. Middle­miss and three other officers were despatched to the scene with Middlemiss as senior officer in charge. Middlemiss, with his previous Himalayan experience, was particularly suited for the study of this problem, and it fell to him to compile all the evidence collected from many sources, and as a result he produced another fine memoir (The Kangra Earthquake of 4th April 1905,

Mem. Geol. Surv. India, 38, 1910).

       In this memoir Middlemiss records all the relevant facts collected and then discusses the cause of the earthquake. This particular earthquake was peculiar because there were two foci or epicentral tracts, the larger one, with a maximum intensity of 10 on the Rossi-Forel scale, being at Dharmsala and Kangra in the Kangra Valley, whilst the subsidiary epicentral tract, with a maximum intensity of 8, was in Dehra Dun, with Dehra Dun and Mussoorie as the principal towns. These two foci are situated where the Sub-Himalayan zone of Tertiary forma­tions embays in a north-easterly direction into the Lesser Himalayan zone, leaving between them a south-west-pointing bulge or bastion of Lesser Himalayan formations - that on which Simla stands. Middlemiss points out that these two embayments are the only such along the whole length of the Himalaya as far as this is known and that they must be regarded as places where the Sub-Himalayan belt has not been straightened out along the over-thrust plane- the Main Boundary fault of the Himalaya- that separates it from the Lesser Himalayan belt to the north-east. He mentions other important features connected with the distribution of these two epicentres, and concludes that the earthquake was of deep-seated tectonic origin. This view of the origin (loc. cit. p. 340): implies that the shock was due to a sudden rupture or release of strain occurring among or below the folded sub-Himalayan formations at two places where the strain was specialy great owing to resistances to the well-established forward march of the over­thrusting foot of the Himalayan range and where packing, with consequent arching , may have brought about a certain loss of isostasy.

 

FOSSILS      The important discovery of the Gondwana fossil plant Gangamopteris by Oetling in the Vihi district of Kashmir, and subsequent research by R. D. Oldham, Hayden, Smith Woodward and Seward on this locality or on fossils therefrom had shown the necessity for a thorou gh examination of such an important area, offering as it did the possibility of correlating the great fresh­water Gondwana formation of Peninsular India with the richly fossiliferous marine sedimentary systems of the Himalayan area-two widely distinct geological provinces not till then ever found in juxtaposition. Middlemiss was entrusted with this task and spent thereon the summers of 1908 and 1909. In crossing the Golabgarh pass into Kashmir he found a supremely important section with two fossil-plant horizons containing Gangamopteris and Glossoperis, of species suggesting respectively the Talchir and Karharbari beds of the Peninsula. These beds were overlain in due course by a limestone containing Protoretepora ampla Lonsd. and other fossils identifying it with the Zewan beds of the Vihi section. Middlemiss had made a most important discovery and thereby established the Permo-Carboniferous (Artinskian) age of the lowest beds of the Gondwanas of the Peninsula of India, and placed these beds in their correct position with reference to the marine fossil sequence of Kashmir (Rec. Geol. Surv. India, 37, 296 (1909)), and therefore of the world.

 

REMAPPING KASHMIR An incidental result of this visit to Kashmir was the discovery that the geological map published by Lydekker in his memoir on the geology of Kashmir, Chamba and Khagan (Mem. Geol. Surv. India, 22 (1883), had been made  on too small a scale, and too rapidly, to satisfy modern requirements; and as a result Middlemiss had to re-lay the foundations of our knowledge of the stratigraphy of Kashmir . (See his paper on 'A Revision of the Silurian-Trias Sequence in Kashmir'. Rec. Geol. Surv. India, 40, 206-260 (1910).) In addition he returned to Kashmir in 1910 and rewrote our knowledge of the Pir Panjal, the Outer or Lesser Himalayan range separating Jammu from Kashmir (Rec. Geol. Surv. India, 41, ll5-144 (1911).) These studies enabled him to apply to Kashmir and Jammu his ideas of the 'successive zonal imbrication' of Hazara (loc. cit. p. 136) already mentioned and he succeeded in recognizing three zones in the part of Kashmir visited by him: (1) the Sub-Himalayan and Nummulitic zone combined, (2) the Pir-Panjal or Carbo-Trias zone, and (3) an older zone or zones containing large areas of rocks of Silurian age. These may be compared with his four Hazara zones listed on page 274.

         Although Middlemiss continued work in Kashmir in 1912 and 1913, with a newly recruited officer, H. S. Bion, he did not contribute any further papers on this work before he retired from the Geological Survey of India in 1917 and became Superintendent of the Mineral Survey of Jammu and Kashmir.  Most of his attention in his new appointment was necessarily directed to the discovery and study of mineral deposits of economic value. But before we discuss this it is necessary to point out that in his surveys in Kashmir between 1909 and 1913 Middlemiss cleaned the slate of previous error and wrote on it the foundations of the geology of Kashmir to serve as a source of inspiration to his successors. The first legatee was Bion, who had been inducted into Kashmir geology by Middlemiss in 1912 and 1913 and had nearly completed his description of 'The Fauna of the Agglomerate Series of Kashmir' when he died prematurely in June 1915. Some years later this study was issued as a memoir in the Palaeontologia Indica (n.s, 12, 1--42 (1928)) with an intro­ductory chapter by Middlemiss. The chief interest of this chapter is Middle­miss's views on the Panjal Volcanics consisting of the Agglomeratic Slate below and the Panjal Trap above, with interbedding of trap flows in the upper part of the Agglomeratic Slate. 'On the origin of the Agglomeratic Slate Middlemiss is not dogmatic, but he inclines to the explosive volcanic theory, the fragmental beds containing intercalated beds with the marine fossils described by Bion (principally species of Spirifer and other brachiopods), the age being Carboniferous (Moscovian and Uralian). The generally over­lying Panjal Traps are of basaltic composition, like the Deccan Traps of the Peninsula, but are of much greater age. A point of much interest brought out by Middlemiss is the variable horizon of the Panjal Traps, so that the horizon of the base of the Trap varies from Middle Carboniferous to Permo­Carboniferous, whilst the upper limit of the Trap ranges from Permo-Car­boniferous to as high as the Upper Trias: so that vulcanicity began in one neighbourhood in Middle Carboniferous times and eventually ended else­where in the same neighbourhood at the close of the Middle Trias. Nevertheless Middlemiss regards these Traps as mainly of extrusive origin, a conclusion acceptable to others who have worked in this country.

 

WADIA A man' s work must be judged partly by the extent to which it benefits and inspires those who follow him. By this test Middlemiss stands high. Bion died young, but D. N. Wadia succeeded in due course to the legacy and in a series of good papers and a massive memoir has built well on the foundations of the new geology of Kashmir laid by Middlemiss, and has been awarded by the Geological Society of London the same medal, the Lyell Medal, as was awarded to Middlemiss.

 

MINERALS KASHMIR On joining his new appointment of Superintendent of the 'Mineral Survey of Jammu and Kashmir', Middlemiss's primary duty became the investigation of the mineral deposits of these territories, with the aid of his assistants. His earlier papers on these minerals are in the Records of the Geological Survey of India and relate to aquamarines (with L. J. Prashad), petroleum and lignite. Of these the most important is his paper on the 'Possible Occurrence of Petroleum in Jammu Province', in which he describes a structure (which he calls the Nar-Budhan dome) in the form of 'a natural reservoir of the best type, suitable for the underground concentration and storage of oil' occurring in rocks of age, structure and lithological character identical with those of the neighbouring oil region of the Rawalpindi plateau (Rec. Geol. Surv. India, 49, 191 (1919)). From this paper it is seen that this was not an accidental discovery, but resulted from a deliberate search based on general grounds of possibilities, and after Middlemiss had made a visit to the Khaur field. Middlemiss decides that the crest beds of his dome are almost certainly lower in the series than the crest beds of the Dhulian dome (now producing oil in the Panjab). Middlemiss's dome has not yet been tested by any oil company, as far as is known to me, perhaps because of the absence of seepages, or perhaps from fear of the effects of a fault mapped by Middlemiss. It was as a part of this work that Middlemiss made his measurement of the inclination of the Main Boundary Fault at Kotli noticed in the footnote on page 274.

    

Reports from Kashmir In 1922 his department commenced its own publication under the title of Reports of the Mineral Survey of Jammu and Kashmir, and between 1922 and 1931 no less than nine of these reports issued from Middlemiss's pen, amply

illustrated with maps and sections, describing mineral occurrences of the State examined by him and his assistants, many of the mineral deposits being fresh discoveries by his department. The titles of these reports are given in the Bibliography. Most of the mineral s described occur in association with a remarkable series of elliptical domes of the Great Limestone, occurring in the midst of the younger Tertiary formation (Murrees) of Jammu, and in the case of the largest dome, that of Riasi, faulted down against the Siwaliks. These structures are unique in the whole length of the Himalaya, and to their upheaval, exposing the rocks underlying the Murrees, is due the fact that Jammu has many mineral deposits exposed at the surface that are not known elsewhere along the Himalaya. The Great Limestone was formerly thought to be of Jurassic age, but Middlemiss equates it with his Infra-Triassic limestone of Hazara, and it is now regarded as of Permian or Permo-Carboniferous age. In this limestone are deposits of ores of copper, lead and zinc

and also barytes; and lying above it is Middlemiss's Bauxitic series, and also the Nummulitic coal deposits of Jammu, long known but now more fully disposed, and in addition they differ from all other Indian deposits of coal in their abnormally low volatiles, so that Middlemiss describes them as anthracitic or semi-anthracitic in character. The cause of this is not known, but it is of great interest that the bauxitic deposits that occur on the surface of the Riasi dome at the next horizon stratigraphically below the Nummulitic coals are also abnormally low in their volatiles, namely water. Middlemiss's report on the bauxitic deposits of Jammu Province is perhaps the most interesting and important of these reports . The Jammu bauxites differ from the well-known horizontally described in one of Middlemiss's reports. These coal deposits are peculiar because instead of lying in the usual synclinal basins they are, of course, anti­clinally disposed  and highly hydrated lateritic bauxites of Peninsular India, which often approximate in composition to the tri-hydrate, in having a decided dip, namely that of the other rock formations of the Riasi dome, and in being monohydrated. In addition they are of unusually high density: yet they are also exceptionally low in iron contents, and very high in alumina, 70 to 80 per cent being usual against the 50 to 60 per cent of normal bauxites. The-quantities of ore discovered by Middlemiss and his assistants are large; but when tested, these bauxites prove to be difficult to dissolve in the Bayer process unless ground exceedingly fine; and this is difficult because these bauxites are also abnormally hard.

 

All the minerals of Kashmir suffer from the fact that Kashmir is distant from the main centres of commerce and industry of India, but it is difficult to suppose that the State will not one day benefit from Middlemiss's  work and see its oil structure fully tested and its rich bauxite deposits worked and perhaps smelted by power generated from the coal found in such juxtaposition or more probably from a projected hydro-electric installation on the Chenab river. Summarizing Middlemiss's economic work, one can say that in this he has shown the same enthusiasm, ingenuity and imagination as in his purely scientific surveys.

 

PENINSULA GNEISS We can now turn to the tniddle period of Middlemiss's labours, namely, that spent mainly in the Peninsula of India, dating from December 1893 until his retirement from the Geological Survey in 1917, the later years being punctuated, as already mentioned, by his visits to Kashmir for summer work. Until 1898 Middlemiss's field work lay in the Salem and Coimbatore districts of the Madras Presidency, in Southern India, where it was directed primarily to the study of the mineral resources of these two districts. This work is represented by four small papers, listed in the Bibliography, devoted to the magnesite deposits and associated rocks of the Salem district, and to the corundum deposits of both districts. But Middlemiss also mapped a consider­able tract of both districts for which there is little to show beyond the references in the Director's annual reports. It seems, however, that he projected a memoir upon the geology and petrology of these two districts, and wrote at least a part of it. But the memoir was never completed. It seems probable either that Middlemiss could not form conclusions satisfactory to himself concerning the mutual relations of the components of this varied petrological terrain, or else that his conclusions and speculations thereon were not acceptable to his Director. 

     Fortunately, in 1917, Middlemiss became President of the Geology Section of the Fourth Indian Science Congress held in Bangalore. The venue of the meeting justified him in his choice of theme for his address, 'Complexities of Archaen Geology in India' (Journ. and Proc. Asiatic Soc. Bengal, n.s. 13, cxcv-cciv, (1917)). By this time the Mysore Geological Department had arrived at two main conclusions concerning the Archaean formations of their State, namely, (1) that the major portion of nearly everyone of the rock members of the Dharwar system of Southern India, including the limestones, conglomerates, quartz-iron-ore schists, and even the quartzites, was of igneous origin, the most sedimentary-looking rocks being regarded as highly altered lavas and tufaceous deposits, and (2) that the Dharwar formation, instead of being younger than the gneissic formations upon which it rests in synclinally disposed strips, was the older and had been intruded into by the gneissic rocks.

The Salem and Coimbatore districts of Madras adjoin Mysore, and the Archaean formations stretch from Mysore into Salem and Coimbatore, so that Middlemiss was entitled to express views on these points based upon his reading of the evidence in the field. Frankly, both views were unacceptable to him, and in his address he gives his reasons why; and incidentally he gives us a glimpse of his past work in Salem, where he had studied carefully the relationship of the Dharwars to the Hosur gneiss, as he terms the 'Funda­mental Gneiss' of Salem. On the former point I was able to support Middlemiss in my address to the Geology Section at the Sixth Indian Science Congress in Bombay (Proc. Asiatic Soc. Bengal, 15, n.s. p. clxxxv (1919) and Mem. Geol. Surv. India, 70, 109 (1936)); and later the Mysore Geology Department has abandoned its nearly all-igneous view of the Dharwars (B. Rama Rao, Presidential address to Geology Section, Indian Science Congress, Indore (1936)).

     On the other point, the relative age of the gneiss, as represented by the Hosur gneiss and the Dharwars, Middletniss admitted that the evidence was conflicting (loc. cit. p. cxcviii):

 

Whilst general conclusions that have great weight are in favour of the younger age of the Dharwars, the particular section given above might be held to prove just the contrary. Only, I think, by looking upon the Hosur gneiss as a rock that bas passed through (it may be) several vicissitudes of solidification and plutonic remelting without ever having developed much intrusive motion as regards the formations above, can the above conflicting testimony be harmonized.

 

Middlemiss's reluctance to accept the Dharwars as older than the gneisses upon which they rest is due to the apparent absurdity of supposing that the larger (the gneisses) call be intrusive into the smaller (the Dharwars), in spite of definite evidence of intrusion at various places. In my own address quoted (loc. cit. p. clxxvi) above I attempted to resolve Middlemiss's dilemma with the following words:

To the stratigrapher the age of an igneous rock dates from the time it solidified, and, if the rock has been molten more than once, its age must date from the time of its latest solidification. Thus we see that the 'fundamental gneisses', because they show intrusive relation s towards the Dharwars, must be regarded as stratigraphically younger, but nevertheless they must, in part, represent the older crust-locally modified to a certain extent by assimilation, no doubt-e-on which, we may assume, the Dharwar sediments were deposited and the Dharwar lavas extravasated.

 

Later in this address Middlemiss discusses certain phenomena in the pre­Cambrian rocks of Idar State in North Bombay, and concludes his address with a passage worth quoting as showing that Middlemiss was mentally prepared (see also p. 275) to agree with those who do not regard all granitic rocks as of igneous origin (loc. cit. p. ccii):

Consequently , it seems to me, that in dealing with any rock that appears to be of doubtful igneous or magmatic origin, it is above all nec essary in these days to ascertain in which direction the cycle of change is moving . To put the matter bluntly-an apparent ortho-gneiss with its contempor aneous veins may quite as well be an intensely meta­morpho sed sediment with pegmatites formed in it by 'selective solution' as it may be the extreme, foliated or otherwise modified , representative of a granitic, gabbroid or hybrid abyssal injection.

 

As this address was given in 1917, the year in which Middlemiss retired from the Geological Survey of India, the passage last quoted must be taken as summing up his long period of work on the rocks of the Peninsula .

 

BURMA Returning now to chronological order, we must note that Middlemiss's work in Southern India was terminated 'by his spending the winter of 1899­1900 in a new field, namely, the Southern Shan States and Karenni in Burma. His very interesting Progress Report (General Report of the Geol. Surv. India  1899-1900, pp. 122-153) does not need further comment here as it dealt with what was only a reconnaissance survey, one however of much use to his successors.

 

MADRAS In 1901 Middlemiss returned to the Madras Presidency, but this time to the northern end known as the Vizagapatam Hill Tracts . Here the only maps available were on the scale of 1 inch = 4 miles, and, as these wild hill tracts are densely clothed with forest and but sparsely inhabited, surveying of the accurate type to which Middlemiss was accustomed was impossible. How­ever, during three field seasons he mapped a large section of this inhospitable country on a broad scale, showing long strips of biotite-garnet-gneisses, of the charnockite series, and of the sillimanite-bearing schists known as khondalites. In addition he made some interesting petrographical discoveries. One was of a sapphirine-bearing rock, sapphirine being a very rare mineral that had hitherto been recorded only from Greenland (Rec. Geol. Surv. India, 31, 38 (1904). The sapphirine occurred in association with rocks containing also green spinel, sillimanite, hypersthene and cordierite, the whole assemblage being due to the mingling of an ultra-basic member of the igneous charnockite series with the sedimentary khondalite series (see T . L. Walker and W. H. Collins, Rec. Geol. Surv. India, 36, 1-18 (1908). A second discovery was of nepheline-syenites similar to miaskite from the Urals and to the nepheline­ syenites of Sivamalai in Southern India (T. L. Walker, Rec. Geol. Surv.India, 36, 19-22 (1908). During his survey of these tracts Middlemiss noticed also that the high-level laterite was limited to a fairly constant level, surrounding the hills like a shore-belt, the higher hills behaving like 'islands in the lateritic age', (See General Report, Gen. Surv. India, for 1902-1903, p. 2S (1903),

 

BOMBAY & IDAR In 1908 Middlemiss was placed in charge of a new party of the Geological Survey of India, the Central India Party, a party that worked continuously for many years extending its field of operations to Rajputana and the Bombay Presidency . Middlemiss remained in charge of this party at intervals until 1915. Besides supervising the work of other officers, Middlemiss made his own field of work in Idar State in North Bombay, at the southern end of the Aravalli ranges of Rajputana . This work is represented by a good memoir on 'The Geology of Idar State (Mem. Geol. Surv. India, 44(1), pp. 1-116 (1921). In this memoir we have another example of Middlemiss's detailed and accurate work, devoted this time to Archaean and other pre-Cambrian formations, with careful petrographical studies of the various rock-types . The importance of this accurate work was seen years later when Dr. Heron, one of the officers of his party , who succeeded ultimately to the charge of the Central India, Rajputana and Bombay Party, carried his own surveys southward from Rajputana to the Idar border and discovered that to make Middlemiss's Idar work join up with his own (Heron's) Rajputana work he had to adjust the nomenclature of some of Middlemiss's formations. This proved easy to do without any re-survey of Idar, in spite of the fact that Middlemiss had been working on an isolated tract not connected by modern surveys to Rajputana, and in spite of the fact that Idar proved to be a much more highly metamorphosed portion of the Aravalli Hills than farther north,

 

OLDHAM Throughout Middlemiss's memoirs and reports the reader will encounter much pleasant and graceful writing. It is not inappropriate, therefore, that almost the last product of his pen geologically is a grateful obituary notice of  R. D. Oldham, the man who first introduced Middlemiss to field work in India, in the Simla Hills (Quart. J. Geol. Soc., 93, ciii. (1937).

 

I hope that the account given in the foregoing pages of Middlemiss's far reaching investigations, especially in Extra-Peninsular India - in Hazara, Kashmir, Garhwal and Kumaon - will impress on all who read that with Middlemiss's death we have lost one of the great figures of Indian geology, much of whose work is destined to become classical.

 

LINK TO MIDDLEMISS' BIBLIOGRAPHY