Himalayan Journal vol.04
The Himalayan Journal

Publication year:

Kenneth Mason
    (Professor GIOTTO DAINELLI)
    (N. E. ODELL)
    (Lieut-Colonel R. C. F. SCHOMBERG)
  14. Lhonak, 1930
    (G. B. GOURLAY)
  15. A Journey in Upper Kumaun and Garhwal
    (Lieutenant HUGH ROSE)
  16. Expeditions
  17. In memorium
  18. Notes
  19. Reviews
  20. Correspondence
  21. Club Proceedings


The Origin of " Kangchenjunga "


The Editor,

The Himalayan Journal.

Dear Sir,

At the express desire of His Highness the Maharaja of Sikkim, I crave the hospitality of your valuable columns to attempt to clear the fog that at present surrounds the correct derivation of the name of the third highest mountain in the world, " Kangchenjunga ".

It may not be commonly known to the outside world that the mountain has a special religious sanctity for the Sikkimese Tibetans (as Mount Kailas, in Tibet known as Kang Tesi or Kang Rinpoche, has for the Hindus) and that their scriptures enjoin a special worship of the deity which embodies the eternal snows. A religious festival has been held annually in Kangchenjunga's honour in Sikkim for hundreds of years past. His Highness the Maharaja of Sikkim is always pleased to keep up this age-long tradition by celebrating the worship of the Snowy Range every year on the 15th day of the 7th Tibetan month, when these time-honoured religious dances by Lamas and laymen are held at Gangtok, the capital of the State.

The assertion of Dr. Hara Prasad Shastri, m.a., c.i.e., that the word Kangchenjunga is of Sanskrit origin and means " Golden Thigh " is based solely on information he secured at Khatmandu from the Nepalese. It is however common knowledge that the advent of the Aryan race of Nepalese into the Nepalese hills and that Sanskrit names and derivations in Nepal cannot claim any greater antiquity than the Tibetan scriptures of the famous saint, Padma Sambhawa. Kangchenjunga is essentially a Sikkim mountain and the Sikkimese, by race, are Tibetans living on the southern slopes of the Himalaya. Before its connection with the British Government in India, Sikkim was a Principality of Tibet and the Maharaja of Sikkim's ancestral home was in Tibet proper. His collateral relatives still reside in that country. There is thus little force in the argument that, " Kangchenjunga " being invisible from Tibet proper, its name could not be of Tibetan derivation. The word Kangchenjunga in Tibetan, or to be more exact, in Sikkimese Tibetan dialect, is written as "l^'^'^'VST (Kangchen mdZod-lNga or Kang-chen Zod-nga), meaning: Snow (Rang), Big (Chen), Treasury (Zod), Five (Nga).

The word should be- pronounced as " Kang-chen Zod-nga". Zod being pronouneed as " Z " and not as " J ". The letter " Z " in the plains of Bengal and Bihar is often pronounced as " J and sometimes in Roman spelling is even written as " J ", for instance, as in the word Mojumdar, which is derived from the Persian word Mauzim. It is not therefore a matter of surprise that the ortho- graphists spelled it with a " J " on hearing the name of the mountain pronounced Kangchenjunga in Bengal and Bihar. Later on, the Sanskritists gave it a Sanskrit derivation by ascribing a Hindi meaning to the word Jungha, which in Tibetan is meaningless. The word Nga corresponds to the Sanskrit letter w and sounds like " ng ", as in the English word " song ".

The details of the five treasuries (mdZod-lNga or Zod-nga) of the large snow mountain (Kangchen) may be found in the Tibetan scriptures known as Lama Gongpa Du-pa, in twelve volumes. The first treasury is of salt, the second of gold and turquoise, the third of holy books and wealth, the fourth of military weapons, and the fifth of crops and medicines.

Even in very learned circles, unconnected with Tibet, a very mistaken notion, I fear, prevails that the Tibetans are not given to naming their mountains. I trust that it would not unduly burden this letter to give the Tibetan names of the twenty-one summits which surround the plateau of Tibet, the Roof of the World. They are:

These mountains are guarded by twenty-one deities called Ge-nyen Nyi-shu Tsa-chig. The details of these are to be found also in the Tibetan scriptures, Lama Gong-pa Du-pa.

The Palace, Yours faithfully,

Gasgtok, Sikkim. Lobzang Chhoden, Rai Bahadur,

1st July, 1931. (Private Secretary to His Highness The Maharaja of Sikkim).


The Editor,

The Himalayan Journal.

Dear Sir,

I have seen with some astonishment the light-hearted way in which the Tibetan origin of the name " Kangchenjunga " has been brushed aside. I would support most strongly the argument in the note by you in the last volume of The Himalayan Journal that Tibetans do name mountains. Most mountains are locally named Kang-Chen (big snow) or Kang-Ri (snow mountain), but other special names are also frequently given to very conspicuous peaks.

I once spent six months at Kamba Bzong in Tibet, north of Kangchenjunga, where this mountain was a most prominent feature, there where the people called it Kang-Chen. The peasants here had certainly not been influenced by any Sanskrit-speaking people from India. As you state in your note, although the mountain is actually outside Tibetan territory, the inhabitants of the neighbouring valleys on the Sikkim side are Tibetan by race and language, and like their Tibetan relatives across the border they also use the name Kang-Chen. In many cases peaks known from India have both an Indian (Sanskrit) name and a Tibetan one, e.g., Kailas or Kang Rinpoche, Gurla Mandhata or Nyimo Namgyal.

There may also be an Indian name for Kangchenjunga. May it not be the Nepalese name, Kumbhkaran Langur ? But let us leave Kang-Chen to the Tibetans.

Satna, Yours faithfully,

28th August, 1931. F. M. BaILEY.

Note by Editor.

Rai Bahadur Lobzang Chhoden's letter is of great interest and should be onriolusive. We very much regret that Dr. Hara Prasad Shastri died last year and no other competent Sanskritist seems prepared to argue for " the Golden Thigh " derivation(1). No one has yet found the word a hundred years old in Bengali writings, and I understand that Dr. Suniti Kumar Chatterji definitely considers tho word has been imported into Bengali by English geography very recently. During the year no letters have been received supporting the Sanskrit derivation.

The last part of Rai Bahadur Lobzang Chhodefi's letter is also important. Are these twenty-one names used geographically to-day and, if so, can they be placed on the map ? Mr. J. Van Manen, the Secretary of the Asiatic Society of Bengal, has promised to investigate this point. Many names of mountains derived from the sacred books of Tibet, either in a corrupted or an abbreviated form, aro used geographically by Tibetans to-day, as has been shown by the late Lieut.- Col. Morslvead; but Mr. Van Manen informs me that there is rarely a sufficiently detailed description in the writings for exact identification; local topographical knowledge seems to be essential and our maps of Tibet are not yet good enough to place many of them, especially if they stand north of the Tibetan plateau.

There is also another point that requires elucidation. Seventeen of these summits terminate with the word Kang ; but four end in Lung. Does this last fact controvert the statement made that a mountain cannot end with a district or valley termination ? Or are these four summits named after neighbouring districts as " Makalu " and " Chamlang " appear to be(2) ?

At the last moment, after going to press, I have received the following important communication on the subject from Mr. Van Manen.

(1) See Himalayan Journal, vol. iii, p. 154.

(2) See pages 175, 193.



The Editor,

The Himalayan Journal.

Dear Sib,

Etymology or the science of the derivation of words is an exact science subject to exact rules and not subject to arbitrary guesses. Two of these rules have been formulated by Skeat, the Nestor of English etymology, as follows :-

(1) Mere resemblance of form and apparent connection in sense between languages which have different phonetic laws or no necessary connection are commonly delusions and are not to be regarded.

(2) Observe history and geography; borrowings are due to actual contact.

If the name Kangchen-dzonga has to be explained these rules should be observed. This involves the following enquiries :-

(1) First of all the history of the name in English literature has to be investigated. We have to trace its first occurrence there and date this appearance, and then have to record, again with dates, all variations of spellings from its first appearance to the present time.

(2) We have to note the various explanations given for the name from time to time.
(3) We have to examine the component parts of the name and the component parts of the words proposed in explanation.

After all this has been done valid conclusions may be drawn. As to the history of the name we have not at present available the complete data. The earliest references I have found appear in a rare publication Dorje-ling, by H. V. Bayley, Calcutta, 1838. On p. 9 this book spells Kunching Jinga, and in Appendix A, the Journal kept by H. Chapman, p. viii, there is an entry, dated December 5th, 1836, giving the spelling Kanxching-jinga. For present purposes we may start with Hooker's Himalayan Journals (1854) which spells the name as Kinchinjunga in the text but as Kangchan-junga on the maps. In the same year, 1854, Cunningham, Laddk (p. 57), spelled Kanchinjinga. After that a bewildering variety of spellings occurs. In 1903 Freshfield stated that " the native name Kangchen- junga, meaning literally ' the five treasuries of greatest snpw ', given by the inhabitants of Sikhim to the five loftiest summits in which the range culminates, probably refers to the roof-like character of the peaks..........” Waddell, Among the Himalayas, 1899, gave the same explanation. In 1891 the same Tibetan scholar had already given this explanation in the Journal of the Asiatic Society of Bengal, Vol. LX (page 53). Graham Sandberg also gave the same explanation in his Manual of the SikMm Bhutia Language, 1895. The Tibetan Dictionaries by Jaeschke (1881) and Desgodins (1899) did the same.

Waddell's explanation was embodied in the Imperial Gazetteer of India (1908). In 1849 Brian H. Hodgson published in the Journal of the Asiatic Society of Bengal an article ' On the physical geography of the Himalaya ' (Vol. XVIII, Part II, p. 761). This was republished together with other articles of the great Himalayan scholar at least in 1857 and 1874, and perhaps oftener.

In this paper as well as on its accompanying map Hodgson mentions the " peak of Kangchan " and in the re-issue of 1857 added in a footnote:

" Kang ' snow '; chan ' abounding in ' having ', like the English suffix full in fearful."

Prom that moment the spelling Kangchan (with an a in the second syllable) enters into geographical literature. For instance Markham, Narrative of the Mission of George Bogle to Tibet, etc., 1876, refers in his preface, p. xxxvi, simply to " the Kangchan ", whilst his maps spell Kanchinjinga.

I have not found evidence that before the publication of Hodgson's article the name was ever spelled other than with an i in the second syllable. Nevertheless Hodgson gave the first syllable correct as kang, not kan.

Now Hodgson's explanation was incorrect. In Tibetan there are two nearly similar forms Kangchen and Kangchan. The first means ' great snows ', the second means ' snowy ' or ' having snow '. Hooker's spelling Kangchan on his maps may be due to this early explanation, which is incorrect. Kangohan means simply Himavat or Himalaya and is applied to a region, not to a single mountain. It is one of the poetic names for Tibet as a whole. Kangchen, great snow, is never applied to Tibet as a whole.

In Tibetan books the name appears in the same spelling as that given by the Tibetan dictionaries. There is a little ritual book containing the form of " worship of the God of the Darjeeling district". The Tibetan title is : Rdo gling yul lhahi gsol mchod.

In it I find the following sentence : gangs chen mdzod Ingahi Iho sgo bsrung. Translated this means : Protect the southern gate of Kangchen-dzonga.

The hi at the end of the name constitutes the genitive termination. It should be borne in mind that in Tibetan spelling and pronunciation differ greatly.

In another (printed) Tibetan work called Dkon mchog spyi hdus, containing various rituals, there is a prayer to Kangchen-dzonga (leaf 61, verso), in which the name is given twice in the abbreviated form of dzfinga alone. This shows that " the five treasuries " are intelligible by themselves and are not to be understood as a " leg " or " thigh " which, without context, would have no meaning.

In the second of the two passages the dzonga has the epithet dam chart, ' holy', attached to it, showing again that the dzonga may be described in various ways and need not be " golden" at all.

I owe both these Tibetan references to my friend Karma Samdhon Paul who, at my urgent request, managed to find the two works at the last moment in Darjeeling.

Thus the problem stood when suddenly, I think early in 1930, a Mr. H. Goshal wrote a letter to the Times saying that Kangchen- dzonga was a name derived from Sanskrit, its first part from kanchana, golden, and its second part from jaugha, thigh. Hinc illae laerymae.

In Sanskrit there is a verbal root kan meaning ' to shine'. Derived from it is an adjective kanaka, golden. An extended form of the above root is kanch, also ' to shine '. Derived from this is the substantive and adjective kanchana, gold or golden. This is the first word alluded to in the above quoted letter to the Times.

The other Sanskrit word is jangha, shank (from the ankle to the knee), but in the older use of the language perhaps also the upper part between knee and waist. This word has passed into several Indian vernaculars where its present-day meaning is not quite definite. Some Bengali dictionaries explain it both as thigh and shin or shank. A Punjabi friend tells me that he thinks that in Punjabi it means leg. In the modern vernaculars the word does not seem to be a very common one and its meaning not quite definite.

With the word kanchana several geographical terms have been made up in Sanskrit. We find in the whole of the Mahabharata only the name of a river, Kanchanakshi., We find further in other Sanskrit literature four names of towns, real or mythical, as follows :-





There is further a word which might be applied to a mountain, Kanchananga, golden-bodied.

There are five compounds connected with mountains as follows :-

Kaiichanagiri, Kanchanadri and Kanchanachala, three names for Mount Meru, and further, Kanchanas'rngin, having gold peaks, and Kanchanavapra, a hill or mound of gold.

Another word containing a Sanskrit form connected with the word for gold is the old city name of Kanchipura, modern Conjeeveram.

In the whole of the accessible mass of Sanskrit lexicographical literature no compound kanchana-jaiigha is on record.

Further, it is not on record that any mountain name should exhibit a descriptive element meaning thigh, shank, or leg. I find that there are many names that exhibit the element peak, point, top, head. We have therefore in the case of the proposed Sanskrit derivation to assign a meaning to the name which does not explain it whereas the Tibetan explanation is a logical one and thoroughly in consonance with the local mystical traditions concerning the mountain.

Now if the Sanskrit derivation were valid it stands to reason that the Sanskritic Indian vernaculars should all of necessity preserve the two elements contained in this derivation. If the word were on the contrary a modern adaptation, of the nature of popular etymology, then various vernaculars might have contrived to make different adaptations. The latter proves to be a fact and this fact is damning to the theory of a Sanskrit derivation. In Mahrati the name for Kangchen-dzonga is not Kanchanjangha but Kanchanganga. The first of these two forms is entirely unknown in Mahrati.

At my request Prof. D. R. Bhandarkar has translated for me the entry under Kanchanganga in the Mahrati Encyclopaedia edited by S. V. Ketkar (1924). Professor Bhandarkar's mother-tongue is Mahrati. The importance of this entry lies not so much in the fact that the author explains the name as a Tibetan word, which he has quite probably taken over from the Imperial Gazetteer of India, but in the fact that it shows that in another vernacular than Bengali the last part of the word Kangchen-dzonga has been Sanskritised in an entirely different way from the way this has been done in Bengali. This argument seems to be clinching. This conclusive argument is confirmed by equally strong evidence from Bengali.

In the Encyclopaedia Bengalica (Visvakosha), (written in Bengali) by Nagendra Nath Vasu and others, Calcutta, from 1885 onwards (Bengali Samvat 1292), there occurs a most significant passage concerning the name under the entry "Kanchanjangha. After a description of the mountain a sentence closes the entry of which the following is the literal translation :-

" At sunrise the peak looks as if golden-and perhaps that is the reason why it is called Kanchanjangha, or Kanchanjingha or Kan- chansrnga, and according to some Sanskrit books Kanchanadri."

There could hardly be more conclusive evidence that the author here contrasts his three alternative Bengali forms with a single different Sanskrit one, and the latter a name which the Sanskrit dictionaries give as a name for Mount Meru and not for Kangchen- dzonga.

For the rest I have asked Dr. S. K. Chatterji, an authoritative Bengali philologist and phonetician, to write a note on this subject which he has kindly done with great precision, learning, and in great detail. This note I communicate herewith. I have further submitted this note to three capable Bengali and Sanskrit scholars who entirely endorse it. They are :-

Prof. D. R. Bhandarkar, Carmichael Professor of Ancient Indian History and Culture, Calcutta University.

Mr. Amulya Charan Vidyabhusana, for many years Secretary, Bangiya Sahitya Parishad, Professor of Pali, Bengali, etc., Vidyasagar College, and Editor of Pancapus'pa.

Mr. Chintaharan Chakravarti, Joint Honorary Secretary, Bangiya Sahitya Parisad, Professor of Sanskrit.

I. also add the transcript of Professor Bhandarkar's translation of the entry Kanchanganga from the Mahrati Encyclopaedia.

In view of all this material I have no hesitation in stating my conviction that the Sanskrit derivation of the name Kangchen-dzonga has to be given up.

One investigation will also repay further enquiry namely that of the number of mountain names entered on the various maps of Tibet beginning with the element Kangchen. At one time or another I have noted I think about a dozen of these but here I will only refer to the Kangchenjhau immediately near Kangchen-dzonga.

In the above note the combination ch in Tibetan and Sanskrit words represents the single Sanskrit or Tibetan letter c as commonly used in philological transcription, pronounced like ch in choke, churn, chill.

The letter n represents the sound ny in canyon or ni in Spaniard.

The letter n represents the ng sound in song.

The letter r represents a sound more or less like the er in Londoner, onlooker, an unsounded er.

The letter s'represents the sh sound.

The letter o represents the German umlaut sound.

Calcutta, Johan van Manen.

February, 1932.

Extract from the Maharastrlya Jnanakosh, or Mahrati Encyclopaedia, edited by S. V. Ketkar (1924, Poona), Vol. 10, p. 235.

Kanchanganga-The second highest mountain peak in the world. This is a mountain peak in the eastern part of the Himalayan range and situated on the boundary between the two states of Sikkim and Nepal. From the station of Darjeeling the whole of this gigantic mountain can be seen from its foot to its snow-clad top. The summit of mount Kanchanganga is 28,146 feet high. Kanchanganga is a Tibetan name and signifies " the five groups of immense heaps of snow ".

Translated by Prof. D. R. Bhandarkar, Carmichael Professor of Ancient Indian History and Culture, Calcutta University.

' Kanchenjunga '

The name I do not consider to be of Sanskrit origin. Like a great many non-Sanskritic geographical names in tracts in and around India, this is the result of Sanskritisation-I should say unconscious Sanskritisation. Everybody in N. Bengal says Tista, which is of Tibeto-Burman origin, but an educated Hindu transformation of the word is Tri-srotas which is good Sanskrit-from books-for " Three Streams ". The Burmese word Mran-ma has given the Indian form Brahma as the name for Burma, through an intermediate pronunciation Bramma; here the Sanskritisation is from the Burmese direct. But when Siam is rendered by Sydma in Bengali and Hindi, it is the European form of the name that is Sanskritised, and this European (Portuguese > English) form itself is a corrupt rendering of a modern Burmese pronunciation of the old Burmese name for the Shan or Tai people of Siam-Rhwam, now pronounced something like Sham. So there is nothing to wonder at a modification of an English rendering of a Tibetan name into something which is intelligible in Sanskrit. Sanskrit is the speech of the Gods-and the Himalaya is the abode of the Gods ; it would seem only natural for any Hindu, and specially for a cultured Hindu, that the names of the peaks of the Himalayas are from Sanskrit. This assumption can only be expected to have the tacit support-and even warm advocacy-from all educated Hindus.

If the word cannot be proved to be from Sanskrit, and if it is really Tibetan (as the evidence seems to prove it conclusively), then there is not much point in pressing for a compromise spelling on a Sanskrit basis. The only ground for retaining the spelling Kanchen- junga (without the g in the first syllable) is that it has got a prescriptive right through so many decades of use, and that in the sentiment of a great many educated Indians the word is Sanskrit, or rather, they would like the name to look like Sanskrit. But Kangchanjunga is the spelling in Hooker, as early as 1854. Personally I think that no educated Indian will break his heart if the first syllable is written Kang: the form Kangchenjunga will still be near enough to his Sanskrit and he will continue to please himself by writing it as a Sanskrit word (Kanchana-janghd) in his Bengali and Hindi and Nepali, and teach it as a Sanskrit name to the village boys in the villages in the plains at the foot of the Himalayas who come to the school to read their geography. Personally I have warned Bengali readers and writers in a note on the transliteration of geographical names which I contributed as a supplement to the biggest and most up-to-date Bengali dictionary (that by Jnanendra Mohan Das) that our Sanskrit name for this high peak is really Tibetan in origin. If the educated Indian really felt about these geographical orthographies, he felt more at Gaurishankar and Deodunga being dethroned by Everest rather than Kanchana-janghd being proved to be, not Sanskrit, and, Aryan (and so associated with the Gods and rishis and heroes of ancient India), but a parvenu in borrowed Sanskrit garb from the neighbouring land of Tibet.

Calcutta. Suniti Kumar Chattekji.


Since writing my note on Kangchen-dzonga I have found an English prototype for the Mahrati form Kanchanganga discussed in that note. It occurs in the form of Kinchingunga on p. 67 of J. Ware Elder's Report on a visit to Sikkim and the Tibetan frontier in October, November and December, 1873. (Calcutta, 1874). It is therefore probable that all vernacular Indian forms except those which are entirely ' learned ' reconstructions (such as Kanchana srnga, ' golden peak', probably is) will ultimately be traced back to definite earlier English prototypes.

I may here also observe that in the Calcutta Amrita Bazar Patrika, dated March 1st, 1932, there is an article about the Dyhrenfurth expedition, containing an interview with Mrs. Dyhrenfurth. The article is to all appearances bodily quoted from the New York Times. Throughout the spelling is Kanchengunga.

One further observation may be recorded. I have come to the conclusion that the name of the mountain has only become generally known in English literature after the publications of Hooker aiid Hodgson, that is about the year 1850.

I am further of opinion that the name has been introduced into European literature through the reports of travellers who learned it on the spot, that is in or near Sikhim, from local inhabitants, and not through Sanskrit literary sources or previously current Aryan vernacular names in India.

Kangchen-dzonga seems indeed a very young geographical name in the West. It is remarkable how rarely this name and the mountain itself are specifically mentioned in the older geographical literature. Early visitors to Darjeeling and Sikhim did evidently not realise that there was one definitely outstanding highest peak amongst the bewildering mass of snow-mountains on the sky line.

A typical example of this is furnished by the ' Narrative Account by Rinzin Nimgyl of his Exploration of the Country to the North and North-West of Kinchinjunga with notes by Col. H. C. B. Tanner and Mr. W. Robert.' This was originally published in the General Report of the Survey of India, 1884-85, (Calcutta, 1886) and subsequently (1915) republished in the Records of the Survey of India, Vol. viii, Part II, p. 359. The mountain itself is not discussed at all in this narrative, and only once quite casually mentioned: "The Jonsong Pass ..........is a continuation of the Kinchinjunga range." The editors mention the name twice in their marginal notes, but in equally casual connections.

In the original, 1886, edition the name is spelled Kinchinjanga ; in the second edition Kinchinjunga. In the first edition the traveller's name was only given by initials.

(In Dyhrenfurth's bibliography the reference to the article is wrong. It is attributed to the Records of the Geological Survey of India; it should be to the Records of the Survey of India. The name of the author is further spelled Nimgyat instead of Nimgyl as in the original).

In the Index to the first eighteen volumes of the Asiatic Researches, 1788-1833, the name does not occur.

The earliest report about a visit to Sikhim which I have traced is a paper by Captain J. D. Herbert ' Particulars of a visit to the Sikhim Hillsetc. (Gleanings in Science, vol. ii, p. 89, Calcutta, 1830). In this paper the name does not occur.

As a matter of fact in earlier literature there is mention of mountains and of snows, but rarely of their names. Besides, the literature is extremely scarce. In Dyhrenfurth's book on the International Himalaya Expedition (Berlin, 1931) there is a very full, though not entirely complete, bibliography on the Kangchen-dzonga region, enumerating 306 items. Amongst these only 3 are dated 1850 or earlier.

Paul Bauer gives in his Im Kampf um den Himalaja, 1931, a bibliography of three and a half printed pages in which no title of any work published before 1850 occurs.

Freshfield, Round Kangchenjunga, 1903, in his lists of books and maps consulted (5 printed pages) only quotes Turner and Hooker as dating from before 1850. Turner, it should be expressly remarked, does not mention the name.

In Gawler's Sikhim, with hints on mountain and jungle warfare, 1873, I have not found the name.

Prom the Index to Puini's II Tibet (Rome, 1904) and the index to Markham's book on Bogle and Manning it would appear that the name was unknown to-is at least not mentioned by-the early Capucin missionaries in Lhasa and the three travellers: van de Putte, Bogle and Manning.

In C. Akanuma's recently published comprehensive Dictionary of Buddhist Proper Names (Nagoya, 1930) a name Kanchanajangha, does not occur.

In d'Anville's Atlas, 7th sheet of Tibet, there is no name resembling Kangchen-dzonga at the approximate place of the mountain, but on the 6th sheet there appears an entry MM Oumoula Kentchong in the left hand lower corner. This is placed due south of Lhasa, 27°N. May this have been meant for Kangchen-dzonga ? This same range appears on the Carte Generale du Thibet in the abbreviated form of Oumoula MM.

Even Hooker does not mention the Kangchen-dzonga in Part II of his Notes of a Tour in the Plains of India, the Himdla, and Borneo, being extracts from, (his) private letters, London, 1849, though his narrative ends with his arrival at Darjeeling.

I now add a few notes on variations in the spelling of the uame.

An unusual spelling which I have noted is the one of Kinchijunga (which may be unintentional) in an article on Darjeeling in the Calcutta Review, vol. xxviii, 1857, p. 203.

Another curious spelling is used in the Gazetteer of Sikhim : Kanchinjingna. It is not a typographical error as it occurs throughout the volume, on pp. 4, 39, 42, 44, 99 and 102. On p. 100 there is, however, a Kanchanjingna, aift on p. 58 a Kanchanjinga.

Colman Macaulay, in his Report of a Mission to Sikhim and the Tibetan Frontier, etc., Calcutta, 1885, writes mostly only Kinchin and rarely, in full, Kinchinjunga.

The Wereschagins in their Reiseskizzen am Indien, Leipzic, 1882 (vol. i, p. 39), mention the name, so far as I have seen, only casually in the title of an illustration and write the " Kantschingaberg " (mount Kantschinga). In the Russian edition of the work, St. Petersburg, 1883 (p. 41), similarly : Gora Kanchinga.

A German version of a work by Paul Mantegazza, Munich, 1921, exhibits the title of Der Kantschindschinga, Tagebuchblatter.

Sven Hedin, in the 2nd German edition of bis Mount Everest, Leipzic, 1926, writes Kantschindschanga.

Sir Richard Temple in his Journals kept in Hyderabad, Kashmir, Sikhim and Nepal, 1887, writes everywhere Kangchanjanga.

Count Goblet d'Alviella, Inde et Himalaya, 1877, spells Kinchin- chinga.

An early derivation of the name is to be found in an article by Capt. W. S. Sherwill entitled ' Notes upon a Tour in the Sikkim Himalayah Mountains, undertaken for the purpose of ascertaining the Geological Formation of Kunchinjinga, etc.' (Journal A. S. B., vol. xxii, 1853, p. 540). He gives the following footnote when first mentioning Kunchinjinga:-

" For the derivation and meaning of this word X am indebted to Lieut. G. B. Mainwaring of the 16th Bengal Grenadiers, who, with a praiseworthy industry, has mastered the Lepcha language, and was, in 1852, engaged upon the study of the Tibetan. The word is Tibetan and means,

English pronunciation. Tibetan equivalents. English.

Kon Khng-s Snow.

Chin Chhn full or covered.

Jong b'jongs Coeval or equal to."

This note is obscure and incorrect. In the Journal the name of the mountain is added below this table in Tibetan [characters but these are misprinted, containing at least three mistakes, and do not correspond to the Tibetan equivalents. The note is worthless.

It is of interest to note that a namesake, Major J. L. Sherwill, published another ' Journal of a trip undertaken to explore the Glaciers of the Kanchunjingah Group in the Sikkim Himalaya, in November, 1861' (J. A. S. B., vol. xxxi, 1862, p. 457). In this later article the name is written throughout as Kanchunjingah.

One further important argument must not be overlooked. Waddell has shown that Kangchen-dzonga as a god, that is the Spirit of the Mountain, is a local Sikhimese god. The Sikhimese are Tibetans by race and language, not Aryans.

Waddell has made a detailed study of the subject. In his Buddhism of Tibet he treats of the matter in full and shows how genuinely local a god Kangchen-dzonga is. The chief reference is on p. 370, and others occur on p. 430 and p. 511.

The same scholar has presented similar material in the section on Lamaism in Sikhim in the Gazetteer of Sikhim. On p. 263 Kangchen-dzfinga is there called " the chief ' country-god of Sikhim'". Facing this page is a picture of the god. On p. 355 the worship of the god is described.

In his more popular book Among the Himalayas, Westminster, 1899, the same author uses the spellings Kanchen and Kanchen- junga, but gives on p. 386 the same explanation of the name as given in the two other works quoted above. On pp. 386-7 he adds explicitly : " The worship of the mountain-god, which dates back to long before the Buddhist period, is celebrated with great pomp every year throughout Sikhim ".

The written and printed Tibetan form Gangs chen mdzod lnga has been quoted before from Tibetan literature.

In Risley's Gazetteer of Sikhim (Calcutta, 1894) there occurs an extract, pp. 1-2, of "a Sikhim paper, which recites various old works." Speaking of the boundaries of the country there is a passage as follows: " The ' mDsod-lNga' mountains and the spirit ' Phra-Man-dGe-Man' of Zar guard it on the north."

This passage shows again that the combination dzdnga is used alone and intelligible as such.

Waddell, The Buddhism of Tibet, London, 1895, summarises on p. 47 seq. the life of Lha-tsiin, the Patron Saint of Sikhim ' as extracted from the local histories'. On p. 49 the name as transcribed from these books is given as Kan-ch'en dso-na, which in our transcription is the same as Kangchen-dzonga.

Sir Charles Bell has an important note in Dyhrenfurth's Himalaya (Berlin, 1931). On pp. 204-5 he states that the Tibetans call the whole of the Kangchen-dzonga group, from Jannu to Narsing ' Tak- tse-Kang i.e. ' Tiger-mountain-snow ', or ' Dzo-nga-Tak-tse i.e. ' Treasury-five-tiger-mountain.' One more striking proof that dzdnga is Tibetan and cannot mean thigh. See also the same author's significant note on p. 203 of the same work, s. v. Kabru, which similarly shows that Kangchen is a Tibetan name and cannot mean golden, whilst the Sikkhimese use it as the name for the peak called Kabru by Europeans. See also the note on p. 209.

In its section on Nomenclature of Places the Gazetteer of Sikhim, p. 44, gives the same spelling and meaning of the name as those endorsed by the Tibetan dictionaries and other authorities quoted in the beginning of my first note.

Concerning popular transformations of foreign names the following two remarks may be made as bearing on our subject.

To what extent Indian Pandits are capable of Sanskritisation is picturesquely shown by the name given to the late Max Muller: Moksha Mula, " root of liberation ", a perfectly correct Sanskrit combination.

In Darjeeling I have myself heard, amongst soldiers, the name Kitchenjungle, not meant as a joke or nickname, but seriously understood as a place where the vegetables for Darjeeling are grown. I am informed that a more common, but obscene, transformation of the name is also current.

It may finally be noted that Sir Charles Bell in his People of Tibet, 1928, in a footnote to the word Kinchinjunga (p. 297) says :

" The pronunciation usual among Europeans. The correct pronunciation is Kang Chen Dzo Nga", " The Five Treasuries of Great Snow."

I have now to close this lengthy note but before doing so I wish, for completeness' sake, to elaborate one point raised before. Skeat enjoins us to observe history when explaining words. For the history of the use of Himalayan names in European literature the early European geographical compilations are of the greatest value. They give clues concerning the dates when specific names have passed into general geographical literature as against their occurrence in reports of restricted circulation and of a specialist character.

An early general and comprehensive work in English is Hugh Murray's Historical Account of Discoveries and Travels in Asia, London, 1820. As far as I have been able to ascertain the name Kangchen- dzonga does not yet appear in it. It does not occur in the Index, or in the map to Vol. ii, nor in the sections dealing with Himalayan travel.

In German Carl Bitter's Die Erdkunde von Asien is a similar work, but on a larger-a gigantic-scale, encyclopedic in nature and containing an enormous number of exact references to early travellers and reports. I have only the second edition at my disposal for reference. In Band III, Part 4, Second Book (Berlin, 1834) the " territory of the Sikhim Raja " is described on p. 104 et seq. I do not find there any reference by name to our mountain. There is mention of " a very high mountain " but the details are vague and not sufficient for identification.

Another encyclopedic and famous German work is Christian Lassen's Indische Alterthumskunde. In Vol. i (2nd edition, Leipzic, 1867) there is on p. 78 (first edition p. 60) a section on Sikhim. It is curious to find that Mount Everest is there placed in Sikhim. Yet there are two passages of interest for us :

" The small principality of Sikkim or more rarely Sikim, which does not play any part in the history of India, has gained this significance for geography that in it gigantic mountains shoulder each other and that the highest mountain on earth rises there. This discovery has been made only recently. The highest mountain is called Koti- vara by the Indians, that is the chief peak, but the English have baptised it with the name Mount Everest in honour of the Surveyor General.............The Kanchinjinga is 28,156 feet high............The Sanskrit name of the Kanchinjinga might be Kanchanajinga. Kanchana means golden and jinga a plant. The shape of the mountain was probably compared to that of this plant. The adjective golden relates undoubtedly to the glow of its snow-covered summit."

We learn two important facts from this extract. First: Sikhim has not played any part in Indian history. Second: a specialist in the old history and Sanskrit literature of India did not know of any classical Sanskrit name for Kangchen-dzonga, and proposed as a probable theory that its second component part might mean, not thigh, but the name of a plant.

Lassen refers in a footnote to an article in Petermann's Mitthei- lungen, 1858, p. 491 et seq., on " the latest English surveys in the central Himalaya ". I have not been able to consult this article.

Calcutta, March, 1932. J.V.M.