The dates of the four papers on this part of the world shew that the first preceded the earliest of the other three by as much as four years; a fact that must be borne in mind when the philological ethnography of New Guinea and the islands to the south and east of it is under notice. The vocabularies of each of the authors illustrated in papers 2 and. 3, more than doubled our previous data—Jukes' illustrating the language of islands between New Guinea and Australia, Macgillivray's those of the Louisiade Archipelago.

That there was a hypothesis at the bottom of No. 1 is evident. Neither is there much doubt as to the fact of that hypothesis being wrong.

I held in 1843 that, all over Oceania, there was an older population of ruder manners, and darker colour than the Malays, the proper Polynesians, and the populations allied to them; that, in proportion as these latter overspread the several islands of their present occupancy the aborigines were driven towards the interior; that in Australia, Tasmania, New Guinea &c. the original black race remained unmolested.

This view led to two presumptions;—both inaccurate;

1. That the ruder tribes were, as such, likely to be Negrito;

2. That the Negrito tongues would be allied to each other.

The view, held by me now, will be given in a future notice.

---

ON THE NEGRITO LANGUAGES.

READ
BEFORE THE PHILOLOGICAL SOCIETY.

FEBRUARY 10, 1843.

By the term Negrito is meant those tribes of the Asiatic and Australian islands, who, in one or more of their physical characters, depart from the type of the nations in their neighbourhood and approach that of the African. The word is more comprehensive than Arafura, Andaman, or Papuan, and less comprehensive than Negro.

Of the Negrito localities the most western are—

The Andaman Islands.—A Vocabulary, collected by Lieutenant R. H. Colebrooke, appears in the Asiatic Researches, vol. iv. p. 410. The native name is Mineopie. An historical notice of them appears as early as the ninth century, in the Travels of the Two Arabians, translated by Renaudot.

The Nicobar and Carnicobar Islands.—In the largest of these it is stated that, in the interior, blacks are to be found. The current assertion concerning the language of the rest of these islands is, that the Carnicobar is Peguan, and the Nicobar Malay.—Asiatic Researches, iii. 303.

Malacca.—The Samangs of the interior are Negrito. For the single Vocabulary of their language, see Crawfurd's Indian Archipelago, or Klaproth's Nouveau Journal Asiatique, xii. 239, where Crawfurd's Vocabulary is reprinted without acknowledgement. The Orang Benua are not Negrito; neither are the Jokong Negrito. For thirty words in the latter language, see Thomas Raffles in Asiatic Researches, xii. 109. In this list twelve words are shown by Raffles to be Malay, and Humboldt states the same of two more. The other sixteen may or may not be of Negrito origin. The Samangs are the Orang Udai.—Humboldt, Über die Kawi-Sprache.

Sumatra.—The Battas of Sumatra are Malay, not Negrito (Marsden's Sumatra, p. 203, and Rienzi's Oceanie, vol. i.). The Sumatran of Parkinson's Journal (p. 198) is the Arabic of Acheen. The true Negritos of Sumatra seem to be,

1. The Orang Cooboo.—These are stated to be pretty numerous between Palembang and Jambee.—Marsden's Sumatra, p. 35.

2. The Orang Googoo,—who are described by the Sumatrans of Laboon as being more Orang Utang than man.—Marsden's Sumatra, p. 35. Specimens of the Orang Googoo (Gougon) Rienzi states to have seen. He says that they come from Palembang and Menangcaboo, and he calls them Pithecomorphi.

For an historical notice as early as 960 A. D., probably referring to the Blacks of Sumatra, see Klaproth in Nouveau Journal Asiatique, xii. 239.

Borneo.—The Biajuk of Borneo is not Negrito but Malay (Crawfurd's Indian Archipelago); neither are the Dyacks Negrito. The statement of Marsden and Leyden is, that the Dyacks are whiter than the rest of the natives of Borneo; and the remark of more than one voyager is, that the Dyacks of Borneo look like South Sea Islanders in the midst of a darker population. Are the Marut, Idongs, Tidongs, or Tirungs of the north of Borneo Negrito? In Rienzi's Oceanie there is a Borneo Vocabulary which is headed Dyack, Marut and Idaan, the three terms being treated as synonyms. Of this Vocabulary all the words are Malay. That there are Negritos in Borneo is most probable, but of their language we possess but one word, apÜn, father[17] (and that more than doubtful); whilst of their name we know nothing; and in respect to their locality, we have only the statement of Kollf, that in the north of Borneo Blacks are to be found on the Keeneebaloo mountain; a statement, however, slightly modified by the fact of his calling them Idaans or Maruts (see Earl's translation of the Voyage of the Doorga, p. 417). Compare the name Idaan in Borneo, with the name Orang Udai, applied to certain rude tribes in Malacca.

The Sooloo Islands.—There are positive statements that the Sooloos contain Negritos. They also contain Malays; as may be seen in a Sooloo vocabulary in Rienzi's Oceanie, vol. i.

The Manillas.—The Isola de Negros testifies its population by its name. Hervas calls it the Papua of the Philippines. In Panay are the blackest of the Philippine Negritos. Rienzi would term them MelanopygmÆi. In Bohol, LeytÉ and Samar, there are Negritos (Lafond Lurcy, ii. 182.); also in Cayagan (Lafond Lurcy, ii. 182.); also in Capul or Abac (Hervas). For the two main islands there are,—1st. In Mindanao, two wild tribes inhabiting the interior, the Bantschilen and the Hillunas. The proof of these two tribes being Negrito is the strongest for the Hillunas. They are the Negros del Monte of the Spaniards (Hervas, Catalogo delle Lingue; Adelung, i. 601). Near MarivÈles are the Igorots or Ætas (Agtas of Hervas); and of these we have late and positive evidence, first to the fact of their being Negrito, and next to the difference of their language from the Tagal.—(Lafond Lurcy.) Secondly, in LuÇon, the Zambalen of Adelung are Negrito. These are the Blacks of Pampango. The Blacks inhabiting the other parts of the island are called Ygelots; and Mount St. Mathew, near Manilla, is one of their well-known localities, and the Illoco mountains another. Here they were visited by Lafond Lurcy. They were all alike, and all under four feet six (French measure). Italonen, Calingas, and Maitim are the names under which the Philippine Blacks have been generally described. Agta and Maitim are said to be indigenous appellations.—Hervas.

Formosa.—The Formosan language is Malay. In the interior, however, are, according to the Chinese accounts,—1, the Thoufan; 2, the Kia-lao; 3, the Chan tchaÓ chan; 4, the Lang Khiao,—aboriginal tribes with Negrito characters, each speaking a peculiar dialect.—Klaproth, Recherches Asiatiques.

The Loochoo Islands.—The current Loochoo language is Japanese (Klaproth, Rech. Asiat.). But besides this, Adelung mentions from PÈre Gaubil and Gosier, that three other languages are spoken in the interior, neither Japanese nor Chinese; and we are now, perhaps, justified in considering that, in these quarters, the fact of a language being aboriginal, is prim facie evidence of its being Negrito.

Java.—Here the evidence of an aboriginal population at all is equivocal, and that of Negrito aborigines wholly absent. For the Kalangs, see Raffles's History of Java. The dark complexions on the island Bali show the darkness, not of the Negrito, but of the Hindoo; such at least is the view of Raffles opposed to that of Adelung (Mith. i.). There is no notice of Blacks in Ende (otherwise Floris), in Sumbawa, or in Sandalwood Island.

Savoo.—If the Savoo of modern geographers be the Pulo Sabatu of Dampier, then there were, in Dampier's time, Blacks in Savoo. The Savoo of Parkinson's Journal is Malay.

Timor.—In this island Negritos were indicated by Peron. Freycinet describes them. Lafond Lurcy had a Timor black as a slave. Of their language he gives four words:—manouc, bird; vavi, woman; lima, five; ampou, ten. All these are Malay.

Ombay.—In Freycinet's Voyage the natives of Ombay are described as having olive-black complexions, flattened noses, thick lips, and long black hair. In Arago[18] we find a short vocabulary, of which a few words are Malay, whilst the rest are unlike anything either in the neighbouring language of Timor (at least as known by Raffles's specimens), or in any other language known to the author. Upon what grounds, unless it be their cannibalism, the Ombaians have been classed with the New Zealanders, is unknown. The evidence is certainly not taken from their language.

Between Timor and New Guinea we collect, either from positive statements or by inference, that, pure or mixed, there are Negritos in at least the following islands:—1, Wetta; 2, Kissa?; 3, Serwatty?; 4, Lette?; 5, Moa?; 6, Roma?; 7, Damma; 8, Lakor?; 9, Luan; 10, Sermatta; 11, Baba; 12, Daai; 13, Serua; 14, the Eastern Arroos; 15, Borassi. (Kollf's Voy.; Earl's Translation.)

The language of the important island of Timor-Laut is Malay. From a conversation with the sailor Forbes, who was on the island for sixteen years, the author learned that there are in Timor-Laut plenty of black slaves, but no black aborigines.

Celebes.—In the centre of Celebes and in the north there are Negritos: the inhabitants call them Turajas, and also Arafuras: they speak a simple dialect and pass for aborigines. (Raffles, History of Java.) Of this language we have no specimen. Gaimard's Menada is the Menadu of Sir Stamford Raffles, and Raffles's Menadu is Malay. (Voyage de l'Astrolabe, Philologie, ii. 191.) The remark made by the collector of this Menadu Vocabulary was, that those who spoke it were whiter than the true Bugis, and that they looked like South-Sea Islanders, a fact of value in a theory of the Dyacks, but of no value in the enumeration of the Negritos.

Bourou, Gammen, Salawatty, Battenta.—For each of these islands we have positive statements as to the existence of Negritos.

Gilolo.—In Lesson's Natural History the inhabitants of Gilolo are classed with those of Gammen, Battenta, &c., as Negritos. The same is the case in the Mithridates, where the inference is, that in all the Moluccas, with the exception of Amboyna and Ternati, Negritos are to be found in the interior. For GuebÉ see the sequel.

The Teetees.—The Teetee Islands of Meares, the Jauts or Aeauw of the Mithridates, sixteen in number, are Negrito. (Meares, Voyage, Adelung.)

Oby.—According to Adelung this island is Negrito.

The object of what has gone before is less to state where Negritos are to be found than where they are to be looked for. Hence many of the above notices indicate the probable rather than the actual presence of them; and those statements concerning the Molucca localities that are taken from systematic books (and as such at secondhand) are all subject to one exception, viz. the fact that the tribes described as Arafura, although in current language Negrito, are not necessarily so. An instance of this has been seen in the so-called Arafura of Menadu. The same applies to the so-called Arafura of Ceram, (Handboek der Land-en Volkenkunde van Nederlandsch IndiË. P. P. Roorda van Eysinga. Amsterdam 1841; indicated by Mr. Garnett,) which is Malay. In the quarters about to be given in detail the evidence is less exceptionable.

New Guinea.—Here there is little except Negritos; and here we meet with the name Papua. What is said of the Papuas must be said with caution. Physical conformation being the evidence, there are in New Guinea two nations, if not more than two:—1. Those of the North, with curly hair, which are subdivided into the pure Papuas, and the Papuas that are looked upon as a cross with the Malay (Quoy, Gaimard and Lesson in the French Voyages). 2. Those of the South, with lank hair, called by the French naturalists Arafuras. The author was unable to determine who were meant by the Alfakis of Quoy (Durville's Voyage, iv. 746). To the language of these Alfakis are possibly referable the ten words of Lesson. These are the numerals, and, they are as might be expected, Malay. For the South of New Guinea we have not so much as a single vocabulary or a single word.

Waigioo.—The Waigioo and New Guinea have been frequently confounded; we have therefore deferred speaking of the latter until we could also deal with the former. Without going into the conflicting evidence, we may state that there are two Vocabularies wherein arm is kapiani, and three wherein arm is bramine. Of the first division we have—1st, the Vocabularies of the Uranie and Physicienne Corvettes, under Freycinet, in 1817, 1818, 1819, as given in Arago's (the draughtsman's) Narrative, p. 275, English translation; and 2ndly, the Undetermined Vocabulary of Dentrecasteaux. Dentrecasteaux, whilst at Boni in Waigioo, saw some strangers who spoke a language very different from the inhabitants of that island; he considered that they came from New Guinea. Now this language is the Waigioo of Arago[19]; whilst the Waigioo of Dentrecasteaux is the Papua of Arago. Among the Vocabularies of the second class we have Gaimard's Rawak Vocabulary, stated especially (Voyage de l'Astrolabe, Philologie, vol. ii. p. 153.) to have been collected at Rawak in Waigioo in 1818: here arm is bramine. Now a vocabulary (that will soon be mentioned) of the New Guinea Papuan of Port Dorey was collected during the expedition of the Astrolabe by the same naturalist, M. Gaimard. With this vocabulary Gaimard's Rawak coincides, rather than with Arago's Waigioo and Dentrecasteaux's Undetermined Vocabulary. This makes the third vocabulary for these islands. The fourth is Gaimard's Port Dorey Vocabulary (Voyage de l'Astrolabe, Philologie, ii. 146.). The fifth, Dentrecasteaux's (or La BillardiÈre) Waigioo Vocabulary. This represents the same language as those last-mentioned, inasmuch as in it arm is bramine not kapiani. The sixth vocabulary is the Utanata, from Dutch authorities (vide Trans. Geogr. Soc). This akin to the Lobo Vocabulary.—Ibid. The next is Forest's Vocabulary. See Forest's Voyage to New Guinea. Such are the data for New Guinea and Waigioo. Dalrymple's Vocabulary will be noticed in the sequel.

GuebÉ.—The GuebÉ Vocabulary of the Astrolabe (Philologie, ii. 157) is the GuebÉ of Freycinet's Voyage in 1818, when it was collected by Gaimard, The GuebÉ of Arago (under Freycinet) also approaches the GuebÉ of Gaimard. According to D. Durville the GuebÉ is Papuan. The author however considers it Malay, though there was some resemblance to the Papuan, inasmuch as many Malay terms were common to both these dialects.

From New Guinea westward and southward the Negritos are no longer isolated. The following are Negrito Islands, or Negrito Archipelagos:—

1. New Britain; 2. New Hanover; 3. New Ireland; 4. Solomon's Islands; 5. Queen Charlotte's Archipelago; 6. Louisiade Archipelago; 7. Isles of Bougainville; 8. Bouka; 9. New Georgia; 10. Admirality Isles,—York, Sandwich, Portland; 11. Santa Cruz Archipelago; 12. Arsacides; 13. Espiritu Santo, or New Hebrides,—Mallicollo, Erromango, Tanna, Erronan, Annatom; 14. New Caledonia; 15. Warouka, Bligh's and Banks's Island.—Astrolabe. The Ticopian is not Negrito but Polynesian.—Voyage de l'Astrolabe.

Fiji Islands.—In the Fiji Islands the physical character of the natives is half Negrito and half Polynesian. Here is the Negrito limit to the east; that is, of Negrito tribes as existing at the present moment.

The languages of the list just given are known to us through the following Vocabularies.

New Ireland &c.—Gaimard's Carteret Harbour Vocabulary.—Voyage de l'Astrolabe, Philologie, ii. 143.

Durville's Port Praslin Vocabulary, incorporated with Gaimard's Carteret Bay Vocabulary.—Ibid.

Dalrymple's so-called New Guinea Vocabulary. The word so-called was used because, unless there were natives of New Ireland on the coast of New Guinea, Dalrymple's Vocabulary is a representative of the Papuan. It coincides with those of Durville and Gaimard from New Ireland: it was collected by Schouten and Le Maire. It is also the New Guinea of De Brosses.

Vocabularies of four small islands are given by Dalrymple and De Brosses, viz. of Moses Island, Moa, Hoorn Island, and Cocos Island. These are the vocabularies of Reland (Diss. xi.), referred to by Adelung.

Manicolo.—In Queen Charlotte's Archipelago, or perhaps among the Solomon Islands, lies an island in name resembling one of the New Hebrides. Durville called it Vanikoro, but Captain Dillon assures me that the true name is Manicolo. Of the language spoken here we have a vocabulary collected by Gaimard in three dialects; the Vanikoro, the Tanema, and the Taneanou. Voyage de l'Astrolabe, Philologie, ii. 164.

Mallicollo.—Cook's Island is Mallicollo. A glossary occurs in Cook's Voyages.

Tanna.—A single vocabulary in Cook's Voyages.

New Caledonia.—A short vocabulary in Cook. A longer one in Dentrecasteaux and La BillardiÈre.

Of the Fiji we have a few words by Cook, a long vocabulary by Gaimard (Astrol. Phil. ii. 136), Port regulations, and MS. Scripture translations, which afford us full and sufficient samples of the language. To deal with this as Negrito the Polynesian element must be eliminated.

In the way of Ethnography Madagascar is Asiatic; since its language, as has been known since the time of Reland, is Malay. For this island the evidence of physical character gives two or more races, but the evidence of language only one.

Australia.—In this island we have vocabularies for the following localities: (1.) Murray Island; (2.) Caledon Bay; (3. 4.) Endeavour River; (5.) the Burrah Burrah tribe; (6.) Limestone Creek; (7.) Port Macquarie; (8.) Port Jackson; (9.) Menero Downs; (10.) Jervis Bay; (11.) Hunter's River, vide Threlkeld's Grammar; (12, 13, 14, 15.) Adelaide,—one of these being Teichelmanns and SchÜrmann's Grammar; (16.) Gulf St. Vincent; (17, 18, 19, 20.) King George's Sound; (22.) Grey's Vocabulary; and a few others.

Van Diemen's Land.—Here, as in Australia, everything is Negrito. In the way of Vocabularies, we have for the North,—(1.) Gaimard's Port Dalrymple Vocabulary, taken down from the mouth of a Van Diemen's Land woman at King George's Sound, with an Englishman as an interpreter.—Voy. Astr. Phil. ii. 9. In the South we have (2.) Cook's Vocabulary, collected in Adventure Bay, S. E. of Van Diemen's Land,—nine words. (3.) Dentrecasteaux's, or La BillardiÈre's Vocabulary. (4.) Allan Cunningham's Vocabulary, collected in 1819 at Entrance Island. (5.) Dr. Lhotsky's Vocabulary, derived from Mr. M'Geary, and representing the language of Hobart's Town.—Journ. Geo. Soc. ix. Besides these, there is a Vocabulary procured by Mr. Robert Brown when in Australia. It nearly represents the same state of language as Dentrecasteaux's Vocabulary.

Besides these remarks, another class of facts should be indicated. In the south of Japan, and in the Marianne Isles, there are statements that Blacks have been:—PÈre Cantova (in Duperrey and Freycinet), and Adelung (Mithr. i.). From Rienzi also we learn a statement of LÜtke's, viz. that in Pounipet, one of the Carolines, there are abundance of Blacks at this moment. These may be indigenous. The hypothetical presence of Negritos may account also for certain peculiarities of the Polynesian of the Tonga Islands. There are traces of them in the Navigator's Archipelago. Crozet (see Pritchard's Phys. Hist.) mentions Negritos in New Zealand, and Cook speaks to a tradition of aboriginal Negritos in Tahiti.

Such are the notices of the Oceanic Negritos in respect to their distribution and the amount of evidence afforded by the specimens of their language. The current opinion is, that over a certain area Blacks of a certain race or races were aborigines. This opinion there is no reason to disturb or to refine upon; the general question is as to the unity or the multiplicity of these races; but the more specific object of the present paper is to ascertain how far that question is decided by the comparison of their languages. The safe way is to ascend in the classification, and to begin with determining the uniformity of speech over limited areas, and within natural boundaries. The most convenient locality to begin with is—

New Guinea.—That four out of the seven New Guinea Vocabularies (supposing them to have been collected independently of each other) represent either dialects of one language, or else languages closely allied, appears on the first comparison. These vocabularies are,—a) Gaimard's Rawak; b) Gaimard's Port Dorey; c) Arago's Papua; and d) Dentrecasteaux's Waigioo. To these Forest's Vocabulary (supposing always that his words have not been incorporated in the vocabularies that came after him) approaches more closely than to the other two.

Of the two remaining vocabularies the Lobo comes nearer to Forest than the Utanata does. Neither, however, coincide with Forest, as Forest coincides with the first four: nor yet do they coincide so closely with each other.

Again:

New Ireland.—As far as we have vocabularies for evidence, the language of New Ireland is one.

For the affinities of the dialects of Moa, Moses Island, Cocos Island, Hoorn Island, to those of New Ireland, see Dalrymple's Island Voyages, ad fin. That the differences in Manicolo are those of dialect, may be seen from Gaimard's Vocabulary.

Australia.—That the Australian languages are one, at least in the way that the Indo-European languages are one, is likely from hence-forward to be admitted. Captain Grey's statement upon the subject is to be found in his work upon Australia. His special proof of the unity of the Australian language is amongst the imprinted papers of the Geographical Society. The opinions of Threlkeld and Teichelmann go the same way. The author's own statements are as follows:—

(1.) For the whole round of the coast there is, generally speaking, no vocabulary of sufficient length that, in some word or other, does not coincide with the vocabulary of the nearest point, the language of which is known to us. If it fail to do this it agrees with some of the remoter dialects. Flinder's Carpentarian, compared with the two vocabularies of the Endeavour River, has seventeen words in common. Of these, three (perhaps) four coincide. Eye, meal, C.; meul, E. R.: hair, marra, C.; morye, E. R.: fingers, mingel, C.; mungal bah, E. R.: breast, gummur, C.: coyor, E. R.

Endeavour River.—Two vocabularies.—Compared with the vocabularies generally of Port Jackson, and the parts south and east of Port Jackson:—Eye, meul, E. R.; milla, L. C.: nose, emurda, E. R.; morro, L. C.: ears, mulkah, E. R.; moko, P. Macquarie: hair, morye, E. R.; mundah, B. B.: breast, coyor, E. R.; kowul, P. J.: fingers, mungal bah, E. R.; maranga, B. B.: elbow, yeerwe, E. R.; yongra, Menero Downs: nails, kotke, E. R.; karungun? P. J.: beard, wollar, E. R.; wato, Jervis's Bay; wollak, Port Maquarie. The number of words submitted to comparison was twenty-two.

Menero Downs (Lhotsky), and Adelaide (G. W. Earl).—Thirteen words in common, whereof two coincide.

Adelaide (G. W. Earl) and Gulf St. Vincent (Astrolabe).

Gulf St. Vincent (Astrolabe) and King George's Sound (Nind and Astrolabe); fifty words in common.

(2.) The vocabularies of distant points coincide; out of sixty words in common we have eight coincident.

(3.) The most isolated of the vocabularies; e. g. the Carpentarian, if compared with the remaining vocabularies, taken as a whole, has certain words to be found in different and distant parts of the island.

The following is a notice of certain words coinciding, though taken from dialects far separated:

In the way of grammatical inflection we find indications of the same unity. We find also differences upon which we should be careful against laying too much stress. The inflection of the number is an instance of the difference. In South Australian—tinyara, a boy; tinyarurla, two boys; tinyar-anna, boys. In Western Australia—yago, a woman; yago-mun, women; goolang, a child; goolang-gurrah, children (gurra, many); doorda, a dog; doorda-goodjal, two dogs; doorda boula, many dogs (boula, many). Here there is a difference where we generally find agreement, viz. in the inflectional (or quasi-inflectional) expression of the numbers. The difference, however, is less real than apparent. The Australian is one of those languages (so valuable in general philology) where we find inflections in the act of forming, and that from the agglutination not of affixes, suffixes and prefixes, but of words. In other terms, inflection is evolving itself out of composition. The true view then of different forms for the same idea is not that the inflections are unlike, but that the quasi-inflectional circumlocutions differ from each other in different dialects. There is no inflectional parallel between two men in English and ?????p? in Greek.

Van Diemen's Land, South.—For the south of Van Diemen's Land the language seems radically one. The following is what Cook has in common with Dentrecasteaux (or La BillardiÈre) and Allan Cunningham.

Lhotsky's Vocabulary stands more alone. With the Vocabulary of 1803 and Dentrecasteaux's Vocabulary, it has but three (or two) coincidences:—tongue, mina Lh.; mene, Voc. of 1803: water, lugana, Lh.; lia, Voc. 1803: drink, lugana, Lh.; laina, Voc. 1803. With Allan Cunningham's Vocabulary it has fourteen words in common and three coincident:—nose, minerana, Lh.; meoun, A. C.: tongue, mina, Lh.; mim, A. C.: fire, lope, Lh.; lope. A. C.. Brown and Cunningham coincide a little more than Cunningham and Lhotsky. It is perhaps safe to say, that for the South of Van Diemen's Land the language, as represented by its vocabularies, is radically one.

Van Diemen's Land, North.—In Lhotsky's Vocabulary seven words are marked W, four E, and one S, as being peculiar to the western, eastern and southern parts of the island. One of the four words marked E is found in the Port Dalrymple Vocabulary, being the only word common to the two, e. g. wood, mumanara, E.; moumra, Port Dalrymple. The coincidence of the North and South is as follows:—

About thirty-five words are common to Lhotsky and the Vocabularies of Brown and Dentrecasteaux. From the foregoing observations we may conclude that for the whole of Van Diemen's Land (as far as represented by the Vocabularies) the language is radically one.

Such are the groups as spread over limited areas and confined within natural boundaries. The affinity of speech between different islands is another question.

Preliminary to this we must eliminate the Malay from the Negrito. The full knowledge that this has been done imperfectly invalidates all that we have arrived at; so that, once for all, it may be stated, that what is asserted respecting the amount of words common to two localities is asserted subject to the condition of their being true Negrito and not Malay.

Andaman and Samang.—Few words in common; one coincident, and that borrowed in all probability from a third language.

New Guinea and Waigioo. By Waigioo is meant the Waigioo of Arago, and the Undetermined Vocabulary of Dentrecasteaux. They have about forty words in common, and the following are coincident:—

New Guinea and New Ireland.—Forest and Dalrymple:—fish, een, F.; hissou, D. Mal.: fire, for, F.; eeff, D. Mal.: sand, yean, F., coon, D.: sun, ras, F.; nass, D: star, mak, F.; maemetia, D. Dalrymple and Utanata.—Upwards of twenty-five words in common:—Earth, taar, D.; tiri Mal.; Ut.: eat, nam nam, D.; nemuka, Ut.: tongue, hermangh, D.; mare, Ut. Dalrymple and Lobo.—About thirty words in common:—arms, pongliman, D.; nimango, Ut., Mal: belly, balang, D.; kanborongo, Ut.: tongue, hermangh, D.; kariongo, Ut.

Port Praslin and Carteret Bay (taken together), and Utanata and Lobo (taken together).—For the sake of comparison, the whole of the words that the two (or four) Vocabularies have in common are exhibited, and by their side the equivalents in Latin and in Greek.

With thirty-seven words in common, the two Negrito languages have seventeen coincident; with thirty-seven words in common; the two classical languages have nine coincident. The evidence, therefore, of the affinity of the Papua and New Ireland is stronger than of the Latin and Greek, as determined from identical data.

New Ireland and Manicolo.—The Port-Praslin and Carteret Bay Vocabularies being dealt with as one for New Ireland, and the three dialects being treated as one for Manicolo, we have, out of twenty-eight words in common, the following coinciding:—yes, io, P. P.; io, C. B.; io, Manic.: eye, mata, P. P.; matak, C. B.; mala, maleo, mataeo, Man., Mal.: banana, ounn C. B.; pounha, ounra, ounro, Man., Mal.: canoe, kouan, C. B.; naoure, goia, koure, Manic, Mal.: tooth, ninissai, P. P.; insik, C. B.; indje, Tanean: testes, puen, P. P.; boua bouinini, boua ini, Man.: beard, kam-bissek, C. B. (incam besser, Dalr.); oungoumie, vingoumie, Man., Mal.: breast, boroick, C. B.; berenhenham, Man.; ear, palalignai, P. P.; pralen, C. B.; manbalenhi, Manic.; hair, nihouge, D.; anaoko, Man.

Manicolo and Mallicollo.—Eighteen words in common, the following coincident:—Bread-fruit, baloe, Man.; barabe, Mall.: cocoa-nut, venoure, Man.; naroo, Mall.: eye, mataeo, Man.; maitang, Mall., Mal.: ear, tagnaini, Man.: talingan, Mall., Mal.: bird, menouka, Man.; moero, Mall., Mal.: head, batcha, Man.; basaine, Mall.: hog, boi boi, Man.; brrooas, Mall., Mal.: no, tae, Man.; taep, Mall.: water, ouine, Man.; ergour, Mall.: drink, kanou, nanou, Man.; nooae, Mall.

Mallicollo and Tanna.—Sixteen words in common:—cocoa-nuts, naroo, Mall.; nabooy, Tann.: drink, noaee, Mall.; nooee, Tann., Mal.: eye, maitang, Mall.; manee maiuk, Tann., Mal.: ears, talingan, Mall.; feeneenguk, Tann., Mal.: bird, mÖeroo, Mall.; manoo, Tann., Mal.: hog, brrooas, Mall.; boogas, Tann., Mal.: navel, nemprtong, Mall.; napeerainguk, Tann.: teeth, reebohn, warrewuk, Mall.; raibuk, Tann.; water, ergour, Mall.; namawarain, Tann.: woman, rabin, Mall.; naibraan, Tann., Mal.

Tanna and Mallicollo (taken together) and New Caledonia.—Neither with Mallicollo or Tanna alone, nor with Mallicollo and Tanna taken together, as compared with New Caledonia, do we find more words coincident than the following:—Cocoa-nut, naroo, M.; nabooy, T.; neeoo, N. Cal., Mal.: drink, noaee, M.; nooee, T.; oondoo, N. Cal.: head, noogwanaium, T.; garmoin (Cook), vangue, (L. B.), N. Cal.: yams, oofe, Tann.; oobe, N. Cal., Mal.: yes, eeo, Tann.; elo, N. Cal.: no, taep, Mall.; nda, N. Cal.

Next in order comes the comparison between the Vocabularies of Van Diemen's Land and South Australia.

Port Dalrymple and King Georges Sound (Nind and Astrol.):—Wound, barana, P. D.; bareuk, N.: wood, moumbra, P. D.; pourn, N.: hair, kide, P. D.; kaat, N.: thigh, degagla, P. D.; tawal, N.: kangaroo, taramei, P. D.; taamour, N.: lips, mona, P. D.; mele, K. G. S.: no, poutie, P. D.; poualt, poort, K. G. S.: egg, komeka, P. D.; kierkee, K. G. S.: bone, pnale, P. D.; nouil, K. G. S. (bone of bird used to suck up water) N.: skin, kidna, P. D.; kiao? K. G. S.: two kateboueve, P. D.; kadjen, K. G. S. (N.). Fifty-six words in common.

Port Dalrymple and Gulf St. Vincent.—Mouth, mona, P. D.; tamonde, G. S. V. (a compound word, since taa is mouth, in K. G. S.): drink, kible, P. D.; kawe, G. S. V.: arm, anme, P. D.; aondo (also shoulder), G. S. V.: hawk, gan henen henen, P. D.; nanno, G. S. V.: hunger, tigate, P. D.; takiou, G. S. V.: head, eloura: P. D.; ioullo, G. S. V.: nose, medouer[20], P. D., modla, G. S. V.: bird, iola, pallo, G. S. V.: stone, lenn parenne, P. D.; poure? G. S. V.: foot, dogna, P. D.; tenna, G. S. V.: sun, tegoura[21], P. D.; tendo, G. S. V. Seventy words in common.

Port Dalrymple and Jervis's Bay.—Wound, barana, P. D.; karanra, J. B.: tooth, iane, P. D.; ira, J. B.: skin, kidna, P. D.; bagano, J. B.: foot, dogna, P. D.; tona[22], J. B.: head, eloura, P. D.; hollo, J. B. Fifty-four words in common. What follows is a notice of some miscellaneous coincidences between the Van Diemen's Land and the Australian.

Such is the similarity amongst the Negrito languages, as taken in their geographical sequence, and as divided into three groups. Between the Andaman and Samang there is no visible similarity or coincidence. From New Guinea to New Caledonia there is a series of coincidences; and there is also similarity between the Australian and Van Diemen's Land. But it is far from following that, because languages will form groups when taken in geographical succession, they will also form groups when the sequence or succession shall be interrupted. Tested by another method there is an affinity as follows:

---

Notwithstanding doubtful words certain, it seems that there is evidence of the most unlike of the languages between Waigioo and New Caledonia (inclusive) being not more unlike than the most dissimilar of the Indo-European tongues. That this statement may be enlarged seems probable by the following parallels:—

The author concluded his paper with the following observations:—

1. For all that is known to the contrary, the Negrito tongues of Sumatra, Borneo, Timor, the Moluccas, Formosa and several smaller islands of whose languages we have no specimens; may be in any relation whatever to any other language, and to each other.

2. The Andamanee and Samang may be in any relation to any other Negrito tongue, or to each other, beyond that of mere dialect.

3. The languages hitherto known of New Guinea, New Ireland, the Solomon's Isles, New Caledonia, Tanna, and Mallicollo, are related to each other, at least as the most different languages of the Indo-European tribe are related.

4. The known languages of Australian are related to each other, at least in the same degree.

5. The Van Diemen's Land and Australian are similarly related.

6. Classified in divisions equally general with the Indo-European, the Negrito dialects (as far as they are known by their vocabularies) cannot fall into more than four, and may possibly be reducible to one; the data being up to a certain point sufficient to determine radical affinities, but nowhere sufficient to determine radical differences.

7. The ethnographical division, according to physical conformation, coincides with the ethnographical division according to language, only so far as the former avoids the details of classification. With the minute subdivisions of the French naturalists the latter coincides least.

8. The distinction between the Negritos and the Malays seems less broad when determined by the test of language, than it does when measured by physical conformation.

9. The notion of the hybridism of the Papuas, arising from the view of their physical conformation, is in a degree confirmed by the nature of their language; although even the physical evidence is not absolute, i. e. on a par with that respecting the hybridism of the Griquas and Confusos.

10. With two[24] (if not more) Negrito tribes, whereof the evidence of language is wholly wanting, physiological differences indicate a probability of difference of language, equal to the difference between any two Negrito languages of which we have specimens.

11. Even in the physiological classifications we are far from being sure that the whole number of Negrito tribes has been described.

Note A.

Note B.

ADDENDA AND CORRIGENDA (1859).

Andaman—The Andaman Language is monosyllabic, and allied to the Burmese of the opposite continent.

Nicoaar &c.—The statement that there are Blacks in the Nicobar Islands is inaccurate. The tribes further from the coast are the rudest. In the Nicobar vocabulary of the Voyage of the Galathea (Steen Bille—Galathea's Reise omkring Jorden), the language most especially represented is that of the island Terressa; the words from Nancovry being marked N, and those from Cariecobar C. N. No difference, beyond that of dialect, is recognized as existing between them. At the same time it is, by no means, certain, that every form of speech belonging to the Archipelago is known to us.

Samang &c.—The statement that these are the Orang Udai is inaccurate. For further notice of the Samang see Newbold's Indian Archipelago; a work not known to me when my paper was written. The ethnology of the Orang Benua is fully illustrated in the Journal of the Indian Archipelago. They are all Malay.

Sumatra.—This island gives us certain tribes ruder than others—not blacker; at any rate no Negritos.

The same applies to Borneo; where there is plenty of barbarism but nothing Negrito.

The same to the Sulu Archipelago.

The Manillas.—Specimens of four of the so-called Negrito languages are to be found in Steen Bille's Voyage of the Galathea (Vol. III.); headed, (1) Umiray, (2) St. Miguel; (3) St. Matheo and (4) Dumagat. They evidently belong to the same group as the Tagal.

Formosa and Loocho.—The criticism that applies to Borneo and the Sulu Archipelago applies here.

For Timor, Ombay &c. see the next paper.

The language of the Arru islanders is not mentioned; indeed in 1843 no specimens of their language had been published. Since, however, a good account of them has been given by Windsor Earl. Their language contains much in common with the languages of the islands to the west of them, whilst in physical appearance they approach the Papuans. They present, in short, transitional characters—Journal of Indian Archipelago, and The Papua Races.

New Britain &c.—For Louisiade forms of speech see the next paper but one; for those of New Caledonia &c. see the fourth.

The Fijis.—The language of the Fijis is Polynesian.

Cocos Island.—The vocabulary of the island so-named seems to me to be that of Ticopia; and, as such, anything but Negrito.

In Braim's Australia we find specimens of five Tasmanian forms of speech. The additions to the philology of Australia since 1843 are too numerous to find place in a notice like the present. The fundamental unity of all the languages of that continent is, now, generally recognized.

Of the Micronesian Islanders (natives of the Marianne and Caroline Archipelagos) some tribes are darker than others. They chiefly occupy the coral, as opposed to the volcanic, formations. The same is the case with the supposed Negritos of Polynesia.

---

ON THE GENERAL AFFINITIES
OF THE
LANGUAGES OF THE OCEANIC BLACKS.

APPENDIX TO JUKES'S VOYAGE OF HMS FLY.
1847.

For philological purposes it is convenient to arrange the Blacks of the Asiatic and Oceanic Islands under five divisions.

I. The Blacks of the Andaman Islands.—These are, comparatively speaking, isolated in their geographical position; whilst the portion of the continent nearest to them is inhabited by races speaking a monosyllabic language.

II. The Blacks of the Malay area.—With the exception of Java, all the larger, and many of the smaller Malay Islands, as well as the Peninsula of Malacca, are described as containing, in different proportions, a population which departs from the Malay type, which approaches that of the Negro, which possesses a lower civilization, which generally inhabits the more inaccessible parts of the respective countries, and which wears the appearance of being aboriginal to the true Malay population. These tribes may be called the Blacks of the Malay area.

III. The Papuan Blacks of New Guinea.—Under this head may be arranged the tribes of New Guinea, New Ireland, the New Hebrides, Tanna, Erromango, Annatom, New Caledonia, &c.

IV. The Blacks of Australia.

V. The Tasmanian Blacks or the Blacks of Van Diemen's Land.

I. The Andaman Blacks will not be considered in the present note.

II. With respect to the languages of the Blacks of the Malay area, it may be stated unequivocally, that the dialects of each and every tribe for which a vocabulary has been examined, are Malay.

A. Such is the case with the Samang, Jooroo, and Jokong vocabularies of the Peninsula of Malacca.—See Craufurd's Indian Archipelago, Asiatic Researches, xii. 109, Newbold's British Settlements in Malacca.

B. Such is the case with every vocabulary that has been brought from Sumatra. The particular tribe sufficiently different from the Malay to speak a different language has yet to be found.

C. Such is the case with the eight vocabularies furnished by Mr. Brooke from Borneo; notwithstanding the fact that both the Dyacks and the Biajuks have been described as tribes wilder and more degraded than the Malay: in other words, as tribes on the Negro side of the dominant population.

D. Such is the case with every vocabulary brought from any of the Molucca, Key, Arru, or Timorian Islands whatsoever; no matter how dark may be the complexion, or how abnormal the hair, of the natives who have supplied it.

E. Such is the case with the so-called Arafura vocabularies of Dumont Durville from Celebes, and of Roorda van Eysinga from Amboyna and Ceram.

F. Such is the case with the languages of the Philippine Islands. In no part of the great Malay area has the difference between the higher and lower varieties of the population, been more strongly insisted on, and more accurately explained than here. Yet the testimony of the early Spanish Missionaries, as to the fundamental identity of the Black with the other languages is unanimous; and, to put the matter further beyond doubt, the few words of the Igorot negroes, near MarivÈles, which are supplied by Lafond Luray, who visited them, are Malay also.

Now, on these grounds, and laying the Andaman Islands out of the question, it may be safely predicated, that, until we reach either New Guinea, or Australia, we have no proofs of the existence of any language fundamentally different from the Malay; whatever may be the difference in physical appearance of those who speak it.

III. For New Guinea, and the islands Waigioo, and GuebÉ, I have found only ten short vocabularies, and these only for the north-western districts. One of these, the GuebÉ, of the voyage of the Astrolabe, although dealt with by Mr. Durville as Papuan, is Malay. The rest, without any exception, have a sufficient portion of Malay words to preclude any argument in favour of their belonging to a fresh class of languages. On the other hand, the commercial intercourse between the Papuans and Malays precludes any positive statements as to the existence of a true philological affinity.

From New Guinea, westward and southward, we have for the localities inhabited by the black tribes with curly hair, the following vocabularies.

1. For New Ireland.

A. Gaimard's Carteret Harbour Vocabulary—Voyage de l'Astrolabe, Philologie, ii. 143.

B. Durville's Port Praslin Vocabulary. Ibid.

C. Dalrymple's, so called, New Guinea Vocabulary, collected by Schouten and Le Maire, given also by De Brosses.

2. For Vanikoro—Gaimard's Vocabulary in three dialects, the Vanikoro, the Tanema, and the Taneanou—Voyage de l'Astrolabe, Philologie, ii. 164.

3. Mallicollo—Cook's Vocabulary.

4. Tanna—Ditto. Also a few words marked G. Bennet, in Marsden's Miscellaneous Works.

5. Erromango—a few words by Bennet, in Marsden.

6. Annatom—Ditto.

7. New Caledonia—A short Vocabulary in Cook. A longer one in Dentrecasteaux and La Billardiere.

All these languages, although mutually unintelligible, exhibit words common to one another, common to themselves and the New Guinea, and common to themselves and the Malay. See Transactions of the Philological Society, vol. i. no.[26] 4.

IV. The Blacks of Australia are generally separated by strong lines of demarcation from the Blacks of New Guinea, and from the Malays. Even on the philological side of the question, Marsden has written as follows—"We have rarely met with any negrito language in which many corrupt Polynesian words might not be detected. In those of New Holland or Australia, such a mixture is not found. Among them no foreign terms that connect them with the languages even of other papua or negrito countries can be discovered; with regard to the physical qualities of the natives it is nearly superfluous to state, that they are negritos of the more decided class."—p. 71.

In respect to this statement, I am not aware that any recent philologist has gone over the data as we now have them, with sufficient care to enable him either to verify or to refute it. Nevertheless, the isolation of the Australian languages is a current doctrine.

I believe this doctrine to be incorrect; and I am sure that, in many cases, it is founded on incorrect principles.

Grammatical differences are valued too high; glossarial affinities too low. The relative value of the grammatical and glossarial tests is not constant. It is different for different languages.

In 1844, I stated, at York, that from three true Malay localities, and in three true Malay vocabularies, I had found Australian and Tasmanian and Papuan words, viz:—

1. In the Timboran dialect of the Sumbawan.
2. In the Mangerei dialect of Flores.
3. In the Ombayan of Ombay.

1. Arm = ibarana, Ombay; porene, Pine Gorine dialect of Australia.

2. Hand = ouiue, Ombay; hingue, New Caledonia.

3. Nose = imouni, Ombay; maninya, mandeg, mandeinne, New Caledonia; mena, Van Diemen's Land, western dialect; mini, Mangerei: meoun, muidge, mugui, Macquarie Harbour.

4. Head = imocila, Ombay; moos, (= hair) Darnley Island; moochi, (= hair) Massied; immoos, (= beard) Darnley Islands; eeta moochi, (= beard) Massied.

5. Knee = icici-bouka, Ombay; bowka, boulkay (= forefinger) Darnley Islands.

6. Leg = iraka, Ombay; horag-nata, Jhongworong dialect of the Australian.

7. Bosom = ami, Ombay; naem, Darnley Island.

8. Thigh = itena, Ombay; tinna-mook (= foot) Witouro dialect of Australian. The root, tin, is very general throughout Australia in the sense of foot.

9. Belly = te-kap-ana, Ombay; coopoi, (= navel) Darnley Island.

10. Stars = ipi-berre, Mangarei; bering, birrong, Sydney.

11. Hand = tanaraga, Mangarei; taintu, Timbora; tamira, Sydney.

12. Head = jahÉ, Mangarei; chow, King George's Sound.

13. Stars = kingkong, Timbora; chindy, King George's Sound, Australia.

14. Moon = mang'ong, Timbora; meuc, King George's Sound.

15. Sun = ingkong, Timbora; coing, Sydney.

16. Blood = kero, Timbora; gnoorong, Cowagary dialect of Australia.

17. Head = kokore, Timbora; gogorrah, Cowagary.

18. Fish = appi, Mangarei; wapi, Darnley Island.

Now as the three dialects have all undoubted Malay affinities, the statement of Marsden must be received with qualifications.

V. Concerning the language of Van Diemen's Land; I venture upon the following statements, the proofs which I hope, ere long, to exhibit in extenso.

a. The Language is fundamentally the same for the whole island; although spoken in not less than four dialects mutually unintelligible.

. It has affinities with the Australian.

?. It has affinities with the New Caledonian.

A fourth proposition concerning the Tasmanian language exhibits an impression, rather than a deliberate opinion. Should it, however, be confirmed by future researches it will at once explain the points of physical contrast between the Tasmanian tribes and those of Australia that have so often been insisted on. It is this—that the affinities of language between the Tasmanian and the New Caledonian are stronger than those between the Australian and Tasmanian. This indicates that the stream of population for Van Diemens ran round Australia rather than across it.

The following affinities occur between the vocabularies published in the present volume and the Malay and Monosyllabic dialects; and they are the result of a very partial collation.

1. Blood = mam, Darnley Island; muhum, South Jooroo dialect of Malacca; mau, Anamitic of Cochin China.

2. Nose = peet, Darnley Island; peechi, Massied; pih, Chinese; pi, Kong Chinese.

3. Face = awop aup, Murray Islands; eebu = (head) Cape York, Massied; oopoo = (head) Tahiti; epoo, Sandwich Islands; aopo, Easter Island.

4. Hair = moos, Darnley Island; mooche, Massied; maow, Chinese.

5. Country = gaed; Darnley Island; kaha, Ternati.

6. Black = gooli, Darnley Island; houli, Tongataboo.

7. Hand = tag, Darnley Island; tangh, Madagascar; tong, Jooro; tay, Anamitic. A current Malay root.

8. Fish = wapi, Darnley Island; iba, Poggy Isles off Sumatra. Also in other Malay dialects.

9. Flame, fire = bae, Darnley Island; api, Flores, or Ende; fai, Siamese; ffoo, Kong Chinese.

10. Hair = yal, Massied; eeal, Cape York; yal, Port Lihou; houlou, Tongataboo.

11. Teeth = dang, Massied; danga, Cape York; dang, Port Lihou; dang'eta, Gunong-talu of Celebes; wahang, Menadu; rang, Anamitic.

The evidence upon which I rest my belief of the fundamental unity of the three philological groups of the Malay, Papua, and Australian languages, is, of the sort called cumulative; and it is the only evidence that our present data will afford us.

Believing, however, in such a fundamental unity, the problem to be solved by further researches on the vocabularies from either Torres Strait or the South of New Guinea, is the problem as to the particular quarter from which New Holland was peopled—whether from New Guinea, or from Timor. Such a problem is not beyond the reach of future philologists.

In the fifth volume of Dr. Prichard's valuable work, I find that Mr. Norris has indicated points of likeness between the Australian dialects, and the Tamul languages of Southern India.

Such may be the case. If, however, the statements of those philologists who connect on one side the Tamul, and on the other the Malay, with the Monosyllabic languages, be correct, the two affinities are compatible.

ADDENDA AND CORRIGENDA (1859).

The error of presuming the ruder tribes to be Negrito is apparent in the notice of the Sumatra, and Borneo tribes. They should have no place in a list of Negritos at all.

The gist of the paper lies in the suggestions to break down (1) the lines of demarcation between the Australians, Tasmanians, and Papuans on one side, and the Malays &c. on the other, and (2) those between the Malay and Monosyllabic tongues.

---

REMARKS ON THE VOCABULARIES
OF THE
VOYAGE OF THE RATTLESNAKE.

APPENDIX TO MACGILLIVRAY'S VOYAGE OF THE HMS RATTLESNAKE. 1852.

In the way of comparative philology the most important part of the Grammar of the Australian languages is, generally, the Pronoun. That of the Kowrarega language will, therefore, be the first point investigated.

In the tongues of the Indo-European class the personal pronouns are pre-eminently constant. i. e., they agree in languages which, in many other points, differ. How thoroughly the sound of m runs through the Gothic, Slavonic, and Iranian tongues as the sign of the pronoun of the first person singular, in the oblique cases; how regularly a modification of t, s, or th, appears in such words as tu, s?, thou, &c.! Now this constancy of the Pronoun exists in most languages; but not in an equally palpable and manifest form. It is disguised in several ways. Sometimes, as in the Indo-European tongues, there is one root for the nominative and one for the oblique cases; sometimes the same form, as in the Finlandic, runs through the whole declension; sometimes, as when we say you for thou in English, one number is substituted for another; and sometimes, as when the German says sie for thou, a change of the person is made as well. When languages are known in detail, these complications can be guarded against; but where the tongue is but imperfectly exhibited a special analysis becomes requisite.

Generally, the first person is more constant than the second, and the second than the third; indeed, the third is frequently no true personal pronoun at all, but a demonstrative employed to express the person or thing spoken of as the agent or object to a verb. Now, as there are frequently more demonstratives than one which can be used in a personal sense, two languages may be, in reality, very closely allied, though their personal pronouns of the third person differ. Thus the Latin ego = e??; but the Latin hic and ille by no means correspond in form with ??, a?t??, and ??e????. This must prepare us for not expecting a greater amount of resemblance between the Australian personal pronouns than really exists.

Beginning with the most inconstant of the three pronouns, viz., that of the third person, we find in the Kowrarega the following forms:—

3.

In the two first of these forms the du is no part of the root, but an affix, since the Gudang gives us the simpler forms nue and na. Pale, the dual form, occurs in the Western Australian, the New South Wales, the South Australian, and the Parnkalla as follows: boola, bulo-ara, purl-a, pud-lanbi = they two.

2.

Here the root is limited to the syllable ngi, as shewn not less by the forms ngi-pel, and ngi-tana, than by the simple Gudang ngi = thou.

Ngi, expressive of the second person, is common in Australia: ngi-nnee, ngi-ntoa, ni-nna, ngi-nte = thou, thee, in the W. Australian, N. S. Wales, Parnkalla, and Encounter Bay dialects.

Ngi-pel is probably thou + pair. A priori this is a likely way of forming a dual. As to the reasons a posteriori they are not to be drawn wholly from the Kowrarega tongue itself. Here the word for two is not pel but quassur. But let us look further. The root p-l, or a modification of it, = two in the following dialects; as well as in the Parnkalla and others—pur-laitye, poolette, par-kooloo, bull-a, in the Adelaide, Boraipar, Yak-kumban, and Murrumbidge. That it may stand too for the dual personal pronoun is shewn in the first of these tongues; since in the Adelaide language purla = ye two. Finally, its appearance amongst the pronouns, and its absence amongst the numerals, occurs in the Western Australian. The numeral two is kardura; but the dual pronoun is boala. The same phenomenon would occur in the present English if two circumstances had taken place, viz., if the Anglo-Saxon dual wi-t = we two had been retained up to the present time amongst the pronouns, and the word pair, brace, or couple, had superseded two amongst the numerals.

Lastly, the Western Australian and the Kowrarega so closely agree in the use of the numeral two for the dual pronoun, that each applies it in the same manner. In the third person it stands alone, so that in W. Australian boala, and in Kowrarega pale = they two, just as if in English we said pair or both, instead of they both (he pair); whilst in the second person, the pronoun precedes it, and a compound is formed; just as if in English we translated the Greek sf?? by thou pair or thou both.

1.

Here the plural and dual are represented, not by a modification of the singular, but by a new word; as different from nga as nos is from ego. The tu, of course, is non-radical, the Gudang form being ngai.

Nga, expressive of the first person, is as common as ngi, equivalent to the second. Thus, nga-nya, nga-toa, nga-i, nga-pe = I, me, in the W. Australian, N. S. Wales, Parnkalla, and Encounter Bay dialects.

Now, the difference between the first and second persons being expressed by different modifications (nga, ngi,) of the same root (ng), rather than by separate words, suggests the inquiry as to the original power of that root. It has already been said that, in many languages, the pronoun of the third person is, in origin, a demonstrative. In the Kowrarega it seems as if even the basis of the first and second was the root of the demonstrative also; since, by looking lower down in the list, we find that i-na = this, che-na = that, and nga-du (nga in Gudang) = who. Ina and chena also means here and there, respectively.

The dual form albei reappears in the Yak-kumban dialect of the River Darling where allewa = we two. Arri = us, is also the first syllable in the Western Australian form arlingul = we; or, rather it is ar-lingul in a simpler and less compounded form. In a short specimen of Mr. Eyre's from the head of the Great Australian Bight, the form in a appears in the singular number, ajjo = I and me. The root tana = they, is not illustrated without going as far as the Western Australian of Mr. Eyre. Here, however, we find it in the compound word par-tanna = many. Its original power is probably others; and it is most likely a widely diffused Australian root.

The pronouns in question are compound rather than simple; i. e. instead of nga = me, and ngi = thee, we have nga-tu and ngi-du. What is the import and explanation of this? It may safely be said, that the termination in the Australian is not a termination like the Latin met in ego-met, inasmuch as this last is constant throughout the three persons (ego-met, tute-met, se-met), whereas, the former varies with the pronoun to which it is appended (nga-tu, and ngi-du). I hazard the conjecture that the two forms correspond with the adverbs here and there; so that nga-tu = I here, and ngi-du = thou there, and nu-du = he there. In respect to the juxta-position of the simple forms (ngai, ngi, and nue) of the Gudang with the compound ones (nga-tu, ngi-du, and nu-du) of the Kowrarega, it can be shewn that the same occurs in the Parnkalla of Port Lincoln; where Mr. Eyre gives the double form ngai and nga-ppo each = I or me.

Now, this analysis of the Kowrarega personals has exhibited the evolution of one sort of pronoun out of another, with the addition of certain words expressive of number, the result being no true inflexion but an agglutination or combination of separate words. It has also shewn how the separate elements of such combinations may appear in different forms and with different powers in different dialects of the same language, and different languages of the same class, even where, in the primary and normal signification, they may be wanting in others. The first of these facts is a contribution to the laws of language in general; the second shews that a great amount of apparent difference may be exhibited on the surface of a language which disappears as the analysis proceeds.

In rude languages the Numerals vary with the dialect more than most other words. We can understand this by imagining what the case would be in English if one of our dialects counted things by the brace, another by the pair, and a third by the couple. Nevertheless, if we bear in mind the Greek forms ?a?assa and ?a?atta, we may fairly suppose that the Kowrarega word for two, or quassur, is the same word with the Head of Australian Bight kootera, the Parnkalla kuttara, and the W. Australian kardura, having the same meaning.

The difference, then, between the numerals of the Australian languages—and it is undoubtedly great—is no proof of any fundamental difference of structure or origin. It is just what occurs in the languages of Africa, and, in a still greater degree, in those of America.

The extent to which the numeration is carried is a matter of more importance. Possibly a numeration limited to the first three, four, or five numbers is the effect of intellectual inferiority. It is certainly a cause that continues it. As a measure of ethnological affinity it is unimportant. In America we have, within a limited range of languages, vigesimal systems like the Mexican, and systems limited to the three first units like the Caribb. The difference between a vigesimal and decimal system arises simply from the practice of counting by the fingers and toes collectively, or the fingers alone, being prevalent; whereas the decimal system as opposed to the quinary is referrible to the numeration being extended to both hands, instead of limited to one. Numerations not extending as far as five are generally independent of the fingers in toto. Then as to the names of particular numbers. Two nations may each take the name of the number two from some natural dualism; but they may not take it from the same. For instance, one American Indian may take it from a pair of skates, another from a pair of shoes. If so, the word for two will differ in the two languages, even when the names for skate and shoe agree. All this is supported by real facts, and is no hypothetical illustration; so that the inference from it is, that, in languages where a numeral system is in the process of formation, difference in the names of the numbers is comparatively unimportant.

The extent to which the numerals vary, the extent to which they agree, and the extent to which this variation and agreement are anything but coincident with geographical proximity or distance, may be seen in the following table:—

The Verb now requires notice. In languages in the same stage of development with the Australian the usual analysis, as shewn by the late Mr. Garnett in his masterly papers on the structure of the verb, is as follows: 1. The root. 2. The possessive pronoun. 3. A particle of time—often originally one of place.

A rough illustration of this is the statement that such a word as dormivi = sleep—my—then (or there). To apply this doctrine to the Kowrarega with our present data, is unsafe. Still, I am inclined (notwithstanding some difficulties) to identify the pa of the Present tense with the bu in kai-bu = now, and the n of the preterite with the n of che-na = there.

The double forms of the Past tense (one in n, and another in m) are at present inexplicable. So are the double forms of the Imperative, viz. the one in r, and the one in e. It may, however, be remarked, that wherever the Imperative ends in e, the Preterite has the form in m; thus, pid-e = dig, pid-ema = dug. The only exception is the anomalous form peneingodgi = dived. This prepares the future grammarian for a division of the Kowrarega Verbs into Conjugations.

The last class of words that supply the materials of comment are the Substantives. Herein, the formation of the plural by the addition of le, probably occurs in several of the Australian tongues. I infer this from many of those words which we find in the vocabularies of languages whereof the grammar is unknown, and which are expressive of naturally plural objects ending in li, la, or l.

1. Star (stars)—pur-le, pi-lle, poo-lle, in Parnkalla, Aiawong, and Yak-kumban.

2. Fire (flames)—ka-lla, gad-la, in W. Australian and Parnkalla.

3. Head (hair)—kur-le, Encounter Bay. Here we learn from the forms kar-ga, from the Head of the Great Australian Bight, and ma-kar-ta, from Adelaide, that the l is foreign to the root.

4. Hands—marrow-la in the Molonglo dialect; and contrasted with marra in the Adelaide.

This, however, is merely a conjecture; a conjecture, however, which has a practical bearing. It suggests caution in the comparison of vocabularies; since, by mistaking an inflexion or an affix for a part of the root, we may overlook really existing similarities.

Father Anjello's very brief grammatical sketch of the Limbakarajia language of Port Essington[27] exhibits, as far as it goes, precisely the same principles as Mr. Macgillivray's Kowrarega; indeed, some of the details coincide.

Thus, the Limbakarajia personal pronouns are—

• I = nga-pi.
• Thou = noie.
• He, she, it = gianat.
• We = ngari.
• We two = arguri.
• Ye = noie.
• They = ngalmo.

Here the pi in nga-pi is the po in the Aiawong nga-ppo; the gian in gian-at being, probably, the in in the Kowrarega ina = that, this. Ngalmo, also, is expressly stated to mean many as well as they, a fact which confirms the view taken of tana.

As for the tenses of the verbs, they are evidently no true tenses at all, but merely combinations of the verbal root, and an adverb of time. In Limbakarajia, however, the adverbial element precedes the verbal one. In Kowrarega, however, the equivalent to this adverbial element (probably a simple adverb modified in form so as to amalgamate with its verb, and take the appearance of an inflexion) follows it—a difference of order, sequence, or position, upon which some philologists will, perhaps, lay considerable stress. On the contrary, however, languages exceedingly similar in other respects, may differ in the order of the parts of a term; e. g. the German dialects, throughout, place the article before the noun, and keep it separate: whereas the Scandinavian tongues not only make it follow, but incorporate it with the substantive with which it agrees. Hence, a term which, if modelled on the German fashion, should be hin sol, becomes, in Scandinavian, solen = the sun. And this is but one instance out of many. Finally, I may add that the prefix apa, in the present tense of the verb = cut, is, perhaps, the same affix eipa in the present tense of the Kowrarega verbs.

Another point connected with the comparative philology of Australia is the peculiarity of its phonetic system. The sounds of f and s are frequently wanting. Hence, the presence of either of them in one dialect has been considered as evidence of a wide ethnological difference. Upon this point—in the case of s—the remarks on the sound systems of the Kowrarega and Gudang are important. The statement is, the s of the one dialect becomes ty or tsh (and ch) in the other. Thus the English word breast = susu, Kowrarega; tyu-tyu, Gudang, and the English outrigger float = sarima, Kowrarega; charima, Gudang,—which of these two forms is the older? Probably the Gudang, or the form in ty. If so, the series of changes is remarkable, and by attending to it we may see how sounds previously non-existent may become evolved.

Thus—let the original form for breast be tutu. The first change which takes place is the insertion of the sound of y, making tyu-tyu; upon the same principle which makes certain Englishmen say gyarden, kyind, and skyey, for garden, kind, and sky. The next change is for ty to become tsh. This we find also in English, where picture or pictyoor is pronounced pictshur, &c. This being the change exhibited in the Gudang form tyutyu (pr. choochoo, or nearly so), we have a remarkable phonetic phenomenon, viz. the existence of a compound sound (tsh) wherein s is an element, in a language where s, otherwise than as the element of a compound, is wanting. In other words, we have a sound formed out of s, but not s itself; or (changing the expression still further) we have s in certain combinations, but not uncombined. Let, however, the change proceed, and the initial sound of t be lost. In this case tsh becomes sh. A further change reduces sh to s.

When all this has taken place—and there are many languages wherein the whole process is exhibited—the sound of a hitherto unknown articulation becomes evolved or developed by a natural process of growth, and that in a language where it was previously wanting. The phenomenon, then, of the evolution of new simple sounds should caution us against over-valuing phonetic differences. So should such facts as that of the closely allied dialects of the Gudang and Kowrarega differing from each other by the absence or presence of so important a sound as that of s.

The comparative absence, however, of the sound of s, in Australian, may be further refined on in another way; and it may be urged that it is absent, not because it has never been developed, or called into existence, but because it has ceased to exist. In the Latin of the Augustan age as compared with that of the early Republic, we find the s of words like arbos changed into r (arbor). The old High German, also, and the Icelandic, as compared with the Meso-Gothic, does the same. Still the change only affects certain inflectional syllables, so that the original s being only partially displaced, retains its place in the language, although it occurs in fewer words. In Australian, where it is wanting at all, it is wanting in toto: and this is a reason for believing that its absence is referrible to non-development rather than to displacement. For reasons too lengthy too exhibit, I believe that this latter view is not applicable to Australian; the s, when wanting, being undeveloped. In either case, however, the phonetic differences between particular dialects are the measures of but slight differences.

Now—with these preliminary cautions against the overvaluation of apparent differences—we may compare the new data for the structure of the Kowrarega and Limbakarajia with the received opinions respecting the Australian grammars in general.

These refer them to the class of agglutinate tongues, i. e. tongues wherein the inflections can be shewn to consist of separate words more or less incorporated or amalgamated with the roots which they modify. It may be said that this view is confirmed rather than impugned.

Now, what applies to the Australian grammars applies also to Polynesian and the more highly-developed Malay languages,—such as the Tagala of the Philippines, for instance; and, if such being the case, no difference of principle in respect to their structure separates the Australian from the languages of those two great classes. But the details, it may be said, differ undoubtedly; and this is what we expect. Plural numbers, signs of tense, and other grammatical elements, are evolved by means of the juxtaposition of similar but not identical elements, e. g. one plural may be formed by the affix signifying many; another, by the affix signifying with or conjointly; one preterite may be the root plus a word meaning then; another the root plus a word meaning there. Futures, too, may be equally evolved by the incorporation or juxtaposition of the word meaning after, or the word meaning to-morrow. All this makes the exact coincidence of the details of inflection the exception rather than the rule.

This doctrine goes farther than the mere breaking-down of the lines of demarcation which separate classes of languages like the Australian from classes of languages like the Malayo-Polynesian. It shews how both may be evolved from monosyllabic tongues like the Chinese or Siamese. The proof that such is really the case lies in the similarity of individual words, and consists in comparative tables. It is too lengthy for the present paper, the chief object of which is to bring down the inferences from the undoubtedly great superficial differences between the languages of the parts in question to their proper level.

In respect to the vocabularies, the extent to which the analysis which applies to the grammar applies to the vocables also may be seen in the following instance. The word hand Bijenelumbo and Limbapyu is birgalk. There is also in each language a second form—anbirgalk—wherein the an is non-radical. So, also, is the alk; since we find that armpit = ingamb-alk, shoulder = mundy-alk, and fingers = mong alk. This brings the root = hand to birg. Now this we can find elsewhere by looking for. In the Liverpool dialect, bir-il = hand, and at King George's Sound, peer = nails. The commonest root, = hand in the Australian dialects, is m-r, e. g.

All this differs from the Port Essington terms. Elbow, however, in the dialects there spoken, = waare; and forearm = am-ma-woor; wier, too, = palm in Kowrarega.

To complete the evidence for this latter word being the same as the m-r of the other dialects and languages, it would be necessary to shew, by examples, how the sounds of m and w interchange; and also to shew (by examples, also) how the ideas of elbow, forearm, and hand do so. But as the present remarks are made for the sake of illustrating a method, rather than establishing any particular point, this is not necessary here; a few instances taken from the names of the parts of the human body being sufficient to shew the general distribution of some of the commoner Australian roots, and the more special fact of their existence in the northern dialects:

---

---

---

---

---

---

The Miriam Vocabulary belongs to a different class, viz. the Papuan. It is a dialect of language first made known to us through the Voyage of the Fly, as spoken in the islands Erroob, Maer, and Massied. Admitting this, we collate it with the North Australian tongues, and that, for the sake of contrast rather than comparison. Here, the philologist, from the extent to which the Australian tongues differ from each other, notwithstanding their real affinity, is prepared to find greater differences between an Australian and a Papuan language than, at the first glance, exists. Let us verify this by reference to some words which relate to the human body, and its parts.

Few Australian vocabularies are thus similar—a fact which may be said to prove too much; since it may lead to inference that the so-called Papuan tongue of Torres Strait is really Australian. Nevertheless, although I do not absolutely deny that such is the case, the evidence of the whole body of ethnological facts—e. g. those connected with the moral, intellectual, and physical conformation of the two populations—is against it.

And so is the philology itself, if we go further. The Erroob pronouns are,

all of which are un-Australian.

Are we then to say that all the words of the table just given are borrowed from the Australian by the Papuans, or vice versÂ? No. Some belong to the common source of the two tongues, pit = nose being, probably, such a word; whilst others are the result of subsequent intercourse.

Still, it cannot absolutely be said that the Erroob or Miriam tongue is not Australian also, or vice versÂ. Still less, is it absolutely certain that the former is not transitional between the New Guinea language and the Australian. I believe, however, that it is not so.

The doubts as to the philological position of the Miriam are by no means diminished by reference to the nearest unequivocally Papuan vocabulary, viz. that of Redscar Bay. Here the difference exceeds rather than falls short of our expectations. The most important of the few words which coincide are

On the other hand, the Redscar Bay word for throat, kato, coincides with the Australian karta of the Gudang of Cape York. Again, a complication is introduced by the word buni-mata = eyebrow. Here mata = eye, and, consequently, buni = brow. This root re-appears in the Erroob; but there it means the eyeball, as shewn by the following words from Jukes' Vocabulary.

Probably the truer meaning of the Redscar Bay word is eyeball.

No inference is safer than that which brings the population of the Louisiade Archipelago, so far, at least, as it is represented by the Vocabularies of Brierly Island and Duchateau Island, from the eastern coast of New Guinea. What points beyond were peopled from Louisiade is another question.

For the islands between New Ireland and New Caledonia our data are lamentably scanty; the list consisting of—

1. A short vocabulary from the Solomon Isles.
2. Short ones from Mallicollo.
3. The same from Tanna.
4. Shorter ones still from Erromanga and
5. Annatom.
6. Cook's New Caledonian Vocabulary.
7. La Billardiere's ditto.

The collation of these with the Louisiade has led me to a fact which I little expected. As far as the very scanty data go, they supply the closest resemblance to the Louisiade dialects, from the two New Caledonian vocabularies. Now New Caledonia was noticed in the Appendix to the Voyage of the Fly (vol. ii. p. 318) as apparently having closer philological affinities with Van Diemen's Land, than that country had with Australia; an apparent fact which induced me to write as follows: "A proposition concerning the Tasmanian language exhibits an impression, rather than a deliberate opinion. Should it, however, be confirmed by future researches, it will at once explain the points of physical contrast between the Tasmanian tribes and those of Australia that have so often been insisted on. It is this—that the affinities of language between the Tasmanian and the New Caledonian are stronger than those between the Australian and Tasmanian. This indicates that the stream of population for Van Diemen's Land ran round Australia, rather than across it." Be this as it may, the remark, with our present scanty materials, is, at best, but a suggestion—a suggestion, however, which would account for the physical appearance of the Tasmanian being more New Caledonian than Australian.

The chief point of resemblance between the Louisiade and the New Caledonian is taken from the numerals. In each system there is a prefix, and in each that prefix begins with a labral letter—indeed the wa of New Caledonia and the pahi of Louisiade seem to be the same roots.

The Redscar Bay numerals are equally instructive. They take two forms: one with, one without, the prefix in ow, as recorded by Mr. Macgillivray.

This system of prefix is not peculiar. The Tanna and Mallicollo numerals of Cook are—

Here, although the formations are not exactly regular, the prefixion of an initial syllable is evident. So is the quinary character of the numeration. The prefix itself, however, in the Tanna and Mallicollo is no labial, as in the Louisiade and New Caledonian, but either k or a vowel.

The next fact connected with the Louisiade vocabularies is one of greater interest. Most of the names of the different parts of the body end in da. In the list in question they were marked in italics; so that the proportion they bear to the words not so ending was easily seen. Now it is only the words belonging to this class that thus terminate. Elsewhere the ending da is no commoner than any other.

What does this mean? If we look to such words as mata-da = eyes, sopa-da = lips, maka-da = teeth, and some other naturally plural names, we should infer that it was a sign of number. That this, however, is not the case is shewn by the equivalents to tongue, nose, and other single members where the affix is equally common. What then is its import? The American tongues help us here.

Now in these, and in numerous other American tongues, the prefix is the possessive pronoun; in other words, there is a great number of American languages where the capacity for abstracting the thing possessed from the possessor is so slight as to make it almost impossible to disconnect the noun from its pronoun. I believe, then, the affixes in question have a possessive power; and am not aware that possessive adjuncts thus incorporated have been recognised in any of the languages for these parts; indeed, they are generally considered as American characteristics.

How far does their presence extend? In the New Caledonian vocabulary of La Billardiere we find it. The names of the parts of the body all take an affix, which no other class of words does. This is gha, guai, or ghai, or other similar combination of g with a vowel. In Van Diemen's Land, an important locality, we find the following series of words, which are submitted to the judgment of the reader.

Here the termination na appears elsewhere, as in memana = fight, nabagee-na = sun; but by no means so frequently; nor yet with such an approach to regularity.

Here however, it must not be concealed that the termination ka, or ga, occurs in other words, such as tenal-ga = laugh, tar-ga = cry, teiri-ga = walk, lamunika = see. These, however, are verbs; and it is possible (indeed probable) that the k or g is the same as in the preceding substantives, just as the m in su-m and e?-? is the m in meus, me, and ??. Still, this will not apply throughout; e. g. the words like lalli-ga = kangaroo, para-ka = flower, and others.

Here, the number of other words ending in na is very considerable; so considerable that, if it were not for the cumulative evidence derived from other quarters, it would be doubtful whether the na could legitimately be considered as a possessive affix at all. It may, however, be so even in the present instance.

To these we may add two lists from the Lobo and Utanata dialects of the south-western coast of New Guinea.

Finally, we have the long, and evidently compound forms of the Corio, Colack, and other Australian dialects; long and evidently compound forms which no hypothesis so readily explains as that of the possessive adjunct; a phenomenon which future investigation may shew to be equally Oceanic and American.

NOTES AND ADDENDA.

The vocabularies of the Rattlesnake are (1) Australian, (2) Papuan.

The former were for the parts about Cape York, i. e. the Northernmost part of Australia, and also the part nearest the Papuan area. The Kowrarega was the form of speech best illustrated.

The Papuan vocabularies were for the Louisiade Archipelago; wholly new as data for a very important and interesting area.

The following paper, connected with the remarks on the incorporation of the possessive pronoun with certain substantives, though on an Asiatic language may find place here.

---

ON A ZAZA VOCABULARY.

READ
BEFORE THE PHILOLOGICAL SOCIETY.

MAY THE 23RD.

The following vocabulary is one taken by Dr. H. Sandwith from a Kurd of the Zaza tribe, one of the rudest of the whole Kurd family, and one for which we have no philological specimens.

The meaning of the termination-min has been explained by Pott and RÖdiger in their Kurdische Studien. It is the possessive pronoun of the first person = my = meus = ???, &c.; so that sÈrÈ-min = caput-meum (or mei), and pie-min = pater-meus (or mei).

So little was the Zaza who supplied Dr. Sandwith with the list under notice able to conceive a hand or father, except so far as they were related to himself, or something else, and so essentially concrete rather than abstract were his notions, that he combined the pronoun with the substantive whenever he had a part of the human body or a degree of consanguinity to name. It is difficult to say how far this amalgamation is natural to the uncultivated understanding, i. e. it is difficult to say so on À priori grounds. That the condition of a person applied to for the purpose of making a glossary out of his communications is different from that under which we maintain our ordinary conversation, is evident. Ordinary conversation gives us a certain number of words, and a context as well. A glossary gives us words only, and disappoints the speaker who is familiar with contexts.

If this be true, imperfect contexts, like the combinations pie-min, &c. should be no uncommon occurrences. Nor are they so. They are pre-eminently common in the American languages. Thus in Mr. Wallace's vocabularies from River Uapes the list run thus:—

similar illustrations being found in almost every American glossary.

In his Appendix to Macgillivray's Voyage of the Rattlesnake, the present writer has pointed out instances of this amalgamation in the languages of the Louisiade. He now adds, that he has also found it in some of the samples of the ordinary Gipsy language of England, as he has taken it from the mouth of English Gipsies.

He considers it to be a personal rather than a philological characteristic, certain individuals having a minimum amount of abstracting power, and such individuals being inordinately common amongst the American Indians.

---

ON THE PERSONAL PRONOUNS AND NUMERALS OF THE MALLICOLLO AND ERROMANGO LANGUAGES.

BY THE REVEREND C. ABRAHAM.

COMMUNICATED WITH REMARKS
TO THE PHILOLOGICAL SOCIETY
by Dr. R. G. LATHAM.

April 22. 1853.

Mallicolo or Sesok?

Erromango.

REMARKS.

Since these vocabularies were laid before the Society, a "Journal of a Cruise among the Islands of the Western Pacific," by Capt. J. E. Erskine, R, N., has been published. This shows the sources of the preceding lists; since the bishop of New Zealand accompanied the expedition, and succeeded in taking back with him, on his return; some youths for the purposes of education.

The class to which these vocabularies belong has never been, sufficiently for the purposes of publication, reduced to writing, nor is any member of it known to scholars in general, in respect to its grammatical structure. This, however, will probably not be the case much longer, since Capt. Erskine has placed the materials for the study of the Aneitum (Annatom) language in the hands of Mr. Norriss, who is prepared for its investigation. Neither has the class been wholly neglected. A grammar of the Tanna (an allied language) was drawn up by Mr. Heath, but it has not been published, and is probably lost. Dr. Pritchard, who had seen extracts from it, writes, that it contained a trinal as well as a singular, a dual, and a plural number. The present list elucidates this. The trinal number (so-called) of the Mallicolo is merely the personal pronoun plus the numeral 3; each element being so modified as to give the appearance of an inflection.

The following tables exhibit the numerals of certain other islands in the neighbourhood. They are taken from Captain Erskine's work, in which reference is made to a "Description of the Islands in the Western Pacific Ocean, by A. Cheyne." This has not been examined by the present writer.

Mr. Abraham's Mallicolo represents the same language with the Mallicolo vocabulary of Captain Cook's Voyages, with which it pretty closely agrees.

His Erromango is more peculiar. Sikai = six = the Mallicolo sukai, which is, itself, nearly the sikai = one. The -ring in suku-ring, too, is the Mallicolo rima. This we know, from the analogies of almost all the languages of Polynesia and the Indian Archipelago, to be the word lima = hand. Hence e-rima (Mallicolo), hand, and suku-ring (Erromango) = one hand. The vat in menda-vat is the Mallicolo -bats in e-bats, the Malay am-pat = four. Du-ru is the Mallicolo e-ry, there being in each case a prefixed syllable. The analysis of tesal and saitavan is less clear. Neither is it certain how ngaraodlen = ten. The other numerals are compounds. This, perhaps, is sufficient to show that the difference between the numerals of the Mallicolo and Erromango is a difference of a very superficial kind. So it is with the Tana, Fotuna, and the first Uea specimens. We must always remember that the first syllable is generally a non-radical prefix.

In the Tana of the preceding table, the words for 6, 7, 8, 9, and 10, seem to be merely the words for 1, 2, 3, 4, and 5 repeated, and something of the same kind appears in the first Uea. Perhaps the representation may be imperfect. At any rate the Tanna of Cook's Voyage runs—

The same appears in the Balad of New Caledonia. Now Cooks New Caledonian runs—

The Yengen and Lifu vocabularies are not so different but that the lu and kun of the one = the luk and yen of the other, as well as the lo and kiuu of the second Uea, and the roo and gen of the Balad.

The importance of these non-radical syllables in the numerals has been indicated by the present writer in the appendix to Mr. M'Gillivray's 'Voyage of the Rattlesnake.' There we find several well-selected specimens of the languages of the Louisiade archipelago. The fact of certain affinities between these and the New Caledonian is there indicated. Each has its prefix. In each the prefix is a labial.

Now the Tana and Mallicolo tongues have a prefix also, but this is not a labial. It is rather a vowel or k (guttural or palatal). Here lies a difference—a difference of detail. Yet the same change can now be shown to be within the pale of the New Caledonian itself, as may be seen by comparing par-roo and par-gen (pah-gen?) with he-luk and he-yen.

The change from r to l creates no difficulty. In one of the Tana vocabularies one = li-ti, in another r-eedee.

These points have been gone into for the sake of guarding against such exaggeration of the differences between the languages of the parts in question as the apparent differences in the numerals have a tendency to engender.

---