Art. I.—On the probable origin of the Maori Races.* This paper is the substance of one read before the British Association at Bristol Aug. 31, 1875, with considerable additions.

Transactions and Proceedings of the Royal Society of New Zealand, Volume 8, 1875, Unnumbered Page

Art. I.—On the probable origin of the Maori Races.* This paper is the substance of one read before the British Association at Bristol Aug. 31, 1875, with considerable additions. By W. S. W. Vaux, M.A., Baliol College, Oxford, and F.R.S. [Communicated to the Wellington Philosophical Society by James Hector, M.D., C.M.G., F.R.S.] The question of the origin of the Maori or native race of New Zealand may, it appears to me, be conveniently considered under the three following heads. I.Their own Traditions; which must, however, be accepted with some reservation, not that we have any right to suppose on their part, an intention to deceive, but because the reports given and published as unquestionable, especially by the Missionaries, are likely, in many instances, to represent rather the ideas of the individual persons who have been specially examined, than the assured judgment of the whole nation. II. The Ethnological Connexion and Affinities, real or imaginary, between them and other peoples, as inferred on scientific principles, or from peculiar existing customs, by European scholars. III. The relation, if any, between the Maori language, as traceable during the last hundred years, and those of the inhabitants of other islands in the Pacific Ocean, indicating, as such a connexion if proved, might be

expected to do, the probability that all the islanders were once one people, and possibly, also, derived at some very remote period, from the distant continents of Asia or America. If these separate lines of research can be shewn to be convergent, it is hoped that some conclusions more or less definite, may be obtained as to the real ancestry of the native population of New Zealand. To take, then,— (1.)Their own Traditions. Now, here, it is interesting to see that a very general uniformity prevails among the legends of all the tribes, the testimony they offer, being, for the most part, that their ancestors found their way to New Zealand from the North or North-East, in certain canoes, the names of which have been preserved; there being also, in the island still existing families, who assert of themselves, that they are lineal descendents of the first immigrants. Some of the natives, I should add, however, believe that they came from the Chatham Islands, the land, geographically, the nearest to New Zealand; but, by far the most prevailing tradition is, that their original ancestral home was Hawaiki, a name, the real or probable meaning of which, I shall fully consider in the later pages of this paper. Now, if the genealogies of the chiefs can be trusted (and, certainly,a priori, there seems no reason for doubting their substantial accuracy), the existing population has occupied these islands but little more than 500 years; while, there would, also, at the same time, seem to be no reliable evidence, that there were any other people settled in them previously, although Mr. Colenso* “N. Z. Inst.,” Vol. I. and some other writers, have warmly advocated the view, that there had been an earlier race there, during a period no man can guess how many ages ago. I think it may be further conjectured that the whole number of original comers was not large, a fact, indeed, we should expect, as they had, in any case, to traverse a considerable breadth of ocean before they could reach these islands: moreover, their method of colonization, by separating into different tribes and families (so as to form distinct settlements, at considerable distances apart the one from the other), agreeing as this does with Captain Cook's account of them, 100 years ago, suggests, also, the probability, that occupation was by successive waves of immigrants, no one of these, most likely, having been very numerous. There are many anecdotes in their various legends of their first advent to New Zealand, which seem to me to bear the stamp of truth upon them: thus, the story of their finding, at two different places, a sperm whale

stranded, is just the sort of incident which would be remembered by an unlettered but observant people, while the further statement that one of the chiefs was so pleased with the beauty of the rata, then in full bloom, that he cast aside the red feathers he was wearing in his head-dress, to don, instead, a circlet of its flowers, is a perfectly natural act, but one, too, which would hardly have been thought of, for the purpose of record, had it not actually occurred. This legend has this further value, that it shows that the arrival of the Maoris must have been in the middle of summer, when the rata is in flower. I confess I don't see any reason why this story should not be accepted in favour of the Maori immigrants, just as readily as the notice, in the “Antiquitates Americanæ,” of the number of hours the sun was above the horizon on the shortest day in “Vinland” (i. e. Narraganset) proving as this observation does, the latitude of the country discovered by the old Icelandic voyagers to the West. * “Antiquitates Americanæ” (Rafn) Copenh. 1837, p. 32. Indeed the long persistency of the Maori traditions and their striking similarity, naturally gives much support to our belief in their general truth; while, unless the conviction of their solid foundation had been fully present to the native mind, we could have hardly conceived the possibility of what is nevertheless literally true, that, only about twenty-five years ago, a family at Tauranga fitted out and provisioned a canoe for the purpose of visiting the homes of their ancestors; † Old traditions brought by the few Maoris who first landed in New Zealand would be preserved, and, perhaps, also, accounts of some of their early wars, but the daily incidents of the expanding population in the new country would not be preserved. As a rule, the histories we learn at school are better impressed on our memories, than the historical incidents during our own lifetime. Incidents occurring in savage or uncivilized states of society, must be surrounded by romance, and elaborated by a generalizing mind, before they pass into tradition or literature. and, more than this, that this is not the only instance in which similar schemes have been entertained and in part carried out. To have thought of such a thing at all, especially as they could have had no definite idea of the direction in which they ought to steer their vessel, proves the vitality of their belief in their traditions, and shows withal no little daring and love of adventure; for the smaller canoes, now or recently in use in New Zealand, could hardly have withstood the billows of the ocean, as did the great double canoes described by Tasman in 1642. It is a popular idea that ignorance of writing tells with fatal effect against the preservation of early traditions: yet it is certain that there is scarcely any limit to the power of memory when exercised on one or a

few subjects only. Everyone knows how an illiterate herdsman will recall at once every little peculiarity of each member of a large herd entrusted to his care, though, at first sight, scarcely anything would seem more difficult in individualizing than the ordinary sheep of a large flock. Again, in all the early and rude states of society, abundant songs and tales of the people are found to have been invariably preserved by the people, ages before any form of writing had been invented: while, we know that, in highly civilized India, where letters were, practically, unknown, even three centuries before the Christian era, the whole of the Sanskrit Vědas, as well as many of the most important commentaries on them, were preserved in the memories of members of the different Brahmanic colleges, whose pride was enlisted in the accurate recollection of the most minute modifications of the sounds and letters of individual words. To maintain the absolute invariabity of these Hymns was the business of their life; and their memories were not distracted by attention to anything else. In the case of the New Zealander, while the demands on his powers of memory were infinitely less, we have reason to believe that his Tohungas, or priests, continued constantly repeating these legends, one from the other, and, no doubt, generally, in the same words. * It is mentioned, I think, by Mr. Ellis, that the native chiefs of the Sandwich Islands have preserved the names of their kings from father to son for a hundred successions—which is by no means improbable as it is the most important, if not the only thing they would care to record. It is not necessary to take these tales for more than they are worth, nor do I wish to claim for them a solid historical basis; but I have as much confidence in them as in the early legends of Greece and Rome, some of which, especially in the case of Rome, are now seen to have had a far more real foundation, than the sceptical historians of the early part of this century were willing to admit. There is nothing, indeed, in the nature of the case, against the probability, that the Maori stories do rest on ultimate facts. Many circumstances, and not the least of these, the admitted fact that the New Zealand chiefs (as was the case, also, in other islands) were, even in life, held to have a quasi-superhuman character, have thrown their mythology into inextricable confusion; but, even, allowing the probability that, as suggested before, some local colourings may have been engrafted on the answers given to the first questions propounded to the Maoris by the missionaries or early settlers, † There is a constant tendency, especially, among uneducated but shrewd savages to give that answer to any question which they think the enquirer would like to have. An Eastern or an Irish peasant illustrates, as well as any one else, this remark. Archdeacon Maunsell has further shown that, in Maori, this practice is, as it were, reduced to a system. “In answering a question,” he says, “the answer will always be regulated by the way in which the question is put, i.e. “Kahore i pai? Ae.” “Was he not willing? Yes,”—i.e. “Yes. He was not willing.” If the answer was intended to be affirmative, the speaker would have said, “I pai ano.” (Maunsell, N.Z. Gr. p. 167.) it does not follow that there was no

truth at all at the bottom of the system recorded, or that the natives did not really believe in a Chief God or Creator, Tangaroa or Tane, with so much of religion, as consists in the recognition of the dependency of the human mind on some presumed higher or more powerful Being. * For interesting details on these matters, see Rev. J. F. H. Wohler's Mythology and Traditions of the Maoris, “Trans., N.Z. Inst.,” Vol. VII., pp. 3–53. I have mentioned the traditions current with respect to the first arrival of the Maoris; but they have also others scarcely less important: for instance, the tribes state, universally, that they were once one people, a statement apparently well confirmed by what can even now be seen; moreover, language, in this instance, may be trusted as a faithful witness. The speech of these Islanders is clearly one and the same; and, though some differences of dialect occur (Mr. Colenso makes ten such varieties; Mr. Shortland six), it is certain the differences between the dialects of the North and South Island are not so great as between Yorkshire and Somersetshire at the present time. Nor, does there seem to me, the slightest ground for supposing them autochthones; the more so, that the only plants they originally cultivated are exotics and their only domestic animal not indigenous. These, therefore, if as we suppose, they found these Islands uninhabited, they must have brought with them whencesoever they came. Another argument in favour of the great antiquity in New Zealand of the present people has been urged on the ground of the presumed long time it must have taken the Maoris to manufacture their most valuable ornaments, hatchets, adzes, spear-heads, etc., in Jade, or other hard and costly materials: and, in this argument there certainly would have been some force, were it certain that either these instruments, themselves, or the substances of which they are made, are found only in New Zealand: on the contrary, however, it is now certain that plenty of highly wrought ornaments of a similar character may be met with in other islands of the Pacific; the presumption being, as well as the strong and natural probability, that if the execution of such works be as difficult, or of the ancient date pretended, the Maoris brought them with them when they first settled in New Zealand. I believe, however, as a matter of fact that these implements do not require for their manufacture, anything like the time suggested by Mr. Colenso. Again,

though there is no doubt that the Middle Island is rich in a kind of Jade, or greenstone, which the Maoris prize highly, it is, by no means, the only place where this mineral is found. * I notice that the South or Middle Island is called by the Maoris Te wahi pounamu, which means, I believe, “the country of Jade.” So far as it goes, I should infer from this name, that Jade was found there abundantly by the first comers from the North Island, and that the name was really given as a reply to enquiries addressed to the natives by the first surveyors. Nor, indeed, do I see, in the writings of any of those persons, who, disbelieving the traditions of the natives, consider them of remote antiquity, † Mr. Colenso's argument from what he thinks the remains of hill forts now covered with humus, can only be answered on the spot by practised antiquaries or geologists. if not, autochthones, any proof, whatever, that the present people differ, in any essential respect, whether of manners or habits, from what has been discovered about the presumed elder races. Yet this would seem to be a matter it is incumbent on these theorists to prove. ‡ Mr. A. Thomson's idea that the present Maoris are a “cross” will be noticed presently. Nor, and this is a more important point, in that it connects the Maoris more or less with their famous extinct bird, the Moa, can I say I have any more faith, in the arguments adduced to show that the Moa-hunters, the population that is, who were the chief agents in the destruction of the Dinornis, were themselves of a period so ancient as to have been contemporary, as has been suggested, with the great European mammals of the Post-Pleiocene period; still less that they dwelt in these Islands so long ago that it may well be doubted whether they have any connection with the present people. It is right, however, to add that, on this particular subject, there is no actual agreement among the holders of these extreme views, as some maintain, like Dr. Haast, that the existing Maoris are descendants of these supposed most ancient Moa-hunters, while others deny this.§ For various interesting suggestions, see Mr. Colenso's summary, “Trans. N.Z. Inst.,” Vol. I., pp. 404–7. Now it has, of late years, been held, with tolerable unanimity, by European ethnologists, that the time of man's existence on the earth admits of division into four principal periods, called, respectively, the Palæolithic, Neolithic, Bronze and Iron Ages, and that the two first periods are marked, definitely, by the use, in the first, of rudely chipped flints and stone implements, and, in the second, by that of materials of the same kind, but, generally exhibiting considerable polish and much skilful and elaborate workmanship,—At the same time, it is tolerably certain that these divisions cannot be drawn with a hard and sharp line, the two classes

having, in many instances, overlapped: thus, rude and finely worked specimens have been, occasionally, found even in Europe so placed as to imply their use by the same population simultaneously. On the supposition, however, that the Palæolithic and Neolithic periods are sufficiently well marked for ordinary purposes of comparison, and bearing in mind that rude as well as polished instruments of stone have been abundantly found in New Zealand, it has been assumed that the people who made and used the former and rudest of them must have belonged to those remote periods; in other words, that there must have been here, as it seems most likely there was once in Europe, a race of men contemporary with the Post-Pleiocene or Mammoth Period. The main argument in favour of this theory rests on the further supposition that all the Moa bones are those of birds extinct for ages, a large number of these remains having been met with in close connection with the flint weapons whereby they were probably slain or cut up. Now, if this be so, it has been further not unreasonably urged that the hunters of the birds must have been contemporaneous with the weapons they used. I ought to add that, with the bones, have been also found a great quantity of the shells of their eggs, as well as the ovens in which they were cooked. Now, no doubt, this theory had a certain consistency so long as it was supposed that most of the bones of the Moas had been found at a depth of many feet under the surface soil, implying, as this circumstance, naturally did, a long lapse of time, since the birds themselves were actually alive on the plains of New Zealand: it was, moreover, asserted that the present people have no traditions of the existence of the bird, which they could not possibly have forgotten. On this point, however, it would seem quite sufficient to remark that the absence of any direct allusion to the Moa in the songs or traditions of the Maoris may just as well have arisen from the probable fact that they were really so familiar with the existence of it, that it would naturally have no place in their traditional lore; while, for the same reason, it would have had none of that peculiar fame among the natives, which the discovery of its remains has aroused among European philosophers. While it lived, the abundant relics of it recently met with shew clearly enough that it could not have been at all rare; and when it perished, perhaps, not very long before the present generation, it simply ceased to be talked about. In making this statement, however, I must not be supposed to deny that Moa bones of considerable antiquity do, from time to time, turn up; I only affirm that that they have not remained long enough in the soil to lose all their albumen and to have been thus converted into true fossils.

Two other views have been put forth, each of which has had its adherents and must, therefore, be noticed here: according to the first, it has been suggested that, had any Moas existed in what may be called “Modern” times, some of the earlier European navigators, as for instance, Captain Cook, * Haast, Dr., Moas and Moa-hunters, “Trans. N. Z. Inst.,” Vol. IV., p. 77. who, during his several visits to these islands, spent over 350 days here, must certainly have heard of or recorded them; according to the second, that, if any of them had died in the open plains, all traces of them would have disappeared in a space of time comparatively short, as the bones of even horses or of other large cattle are known not to resist exposure to the weather for more than 20 or 30 years; the inference being, that all the remains which have been discovered, owe their preservation to the silt and mud in which they have been very generally found embedded. It was, besides this, averred by some that so ancient, indeed, was the period of the final disappearance of the Moa, that its life probably preceded more than one great geological change, such as that which created Cook Strait; while the occurrence of many species of Struthia as well of the Dinornis suggested, at least to these reasoners, the probability of a former vast Antarctic continent, † Admiral Dumont D'Urville in the “Voyage of the Astrolabe” has held this view. connectiilg Australia, through New Zealand, with America and perhaps, even with Africa; the whole of this continent, with the exception of Madagascar, Australia, and New Zealand having been since submeyged for countless ages. The smaller groups, more strictly known as Polynesia, weye not, I believe, included in this grand conception, as being when not volcftnic for the most part of coralline formation. But, apart from other considerations, it is clear that such a, speculation requires an enormous amount of evidence to render it at dl probable; nor, indeed, am I aware that it has been in any way confirmed by competent geologists; moreowr, so far as the present enquiry goes, by siiuplifying, it may be, but, in a very doubtful manner, the problem to be discussed, it looks to me very much like cutting the hot, we ouglit rather to atteinpt untying. Of conrse, if there were any real evidence (such as me see in channels so narrow as tlhat between England and France) that, Within a tolerably recent period; and Australia on the one aicie, and with India, via the Malay Peninsula, ctc., on the other, we should have the chance of solving many difficulties, which now beset the enqniry into the origin of many races, whose languages, so fay as they can be trusted, certaidy shew considerable signs of affinity. Late reseascies have, however, shewn, conclusively, that the idea

of the extraordinary antiquity of the last living Moa cannot be maintained; indeed, considering the vehemence with which the antiquarian theory has been urged, it is not a little remarkable that the first to discover its remains (Bishop Williams and the late Rev. R. Taylor in 1839) stated their beliefs at the time that the actual extinction of the bird was quite recent; a view, in which they were subsequently energetically supported by Mr. Walter Mantell (the son of the well-known geologist), himself, at once, by far the greatest collector of Moa remains, and the person to whom, more than to any one else, all the Museums in Europe have been indebted for the specimens they now possess. Elaborate examinations of the districts or individual spots either where bones have been accidentally discovered or theoretically guessed at, as those most likely to prove rich in such relics* See the following papers:— Mantell, W. B. D., “Quar. Jour. Geo. Soc.,” and “Trans. N. Z. Inst.,” Vol. I. Haast, Julius, F.R.S., On Moa Hunters and Moas, “Trans. N. Z. Inst.,” Vol. I., p. 80; Vol. IV., p. 66 and p. 102; Vol. VI., p. 62 and p. 419; Vol. VII., p. 54. Hector, Dr. J., F.R.S., Discoveries of Moa Remains, “Zool. Proc.,” 1867; “Trans. N. Z. Inst.,” Vol. IV., p. 110; Vol. V., p. 407; Vol. VI., p. 76 and p. 370. Murison, W. D., On Moa Remains, “Trans. N. Z. Inst.,” Vol. IV., pp. 124–4. Williams, W. L., On Footprints of a Large Bird, “Trans. N. Z. Inst.,” Vol. IV., pp. 124–7. Mantell, W. B. D., On Moa Beds, “Trans. N. Z. Inst.,” Vol. IV., pp. 94-97. Grey, Sir Geo., On the Hokioi, “Trans. N. Z. Inst.,” Vol. V., p. 435. Taylor, Rev. J. R., On the First Discovery of Moa Remains, “Trans. N. Z. Inst.,” Vol. V., pp. 97-101. Fraser, Captain, On Earnscleugh Cave Remians, “Trans. N. Z. Inst.,” Vol. V., pp. 97-105 Hutton, Captain, F.G.S., “Trans. N. Z. Inst.,” Vol. V., p. 138 and p. 266. M'Kay, Alex., On Moa Hunters, “Trans. N. Z. Inst.,” Vol. VII., p. 98. Booth, J., Moa Swamps, “Trans. N. Z. Inst.,” Vol. VII., p. 123. Hamilton, J. W., Traditions of Moas, “Trans. N. Z. Inst.,” Vol. VII., p. 121. Goodall, J.,—he thinks, quoting Hochstetter, p. 210, that there was an antecedent population—“Trans. N. Z. Inst.,” Vol. VII., p. 144. Stack, Rev. J. W., who holds that there is no evidence of Moa in the N. Z. poems collected by Sir G. Grey in his article, “Trans. N. Z. Inst.,” Vol. IV., No. 5; Append. XXVIII.-IX., Vol. VII. Roberts, W. H. S., Evidence of Modern Arts, “Trans. N. Z. Inst.,” Vol. VII., p. 548. have since been carried out and discovered by Drs. Haast, Hector, Captain Hutton, Messrs. Murison. Booth, and others, the result, being, the discovery of many portions of these birds, with not only their skin and muscles adhering to their bones, but even their feathers, a combinatioi which could not have been preserved, had not the destruction of these individual birds, at least, been quite modern. In the course of these researches, I may add,

extensive and long occupied camping grounds of the Moa-hunters have been explored, and many essential points relative to their mode of life have been ascertained. By these means it has been shewn, that the argument for their remote antiquity, derived from the discovery of rude as well as highly finished stone implements, falls to the ground. Clearly they are found thus intermixed, as having been in use at one and the same time by the same people. As both these classes, together with a large quantity of flakes of flint and chert have been discovered in these camping grounds, it is reasonable to suppose that each type of implement was used just as might be required for the matter in hand. If, for instance, the Moa-hunters wished to grub up fern roots, they would use, as the Maoris do now, the hard and often highly polished adzes; if they were desirous of stripping the flesh off the tough neck of the bird, they would make use of the sharp flakes of flint; if, lastly, they were anxious to break the thick bones of the tibia, to get at the marrow, they would use the rough and massive kapus, as suggested by Dr. Hector. The fact is, no satisfactory reasons can be deduced for the age of those, who used these weapons, merely from the circumstances under which they have been found; but, at the same time, there is clearly no proof of their remote antiquity; moreover, what may be called a manufactory of flint flakes is constantly associated with the Moa bones where most abundant. Again, recurring to the idea of the ignorance of the natives, as inferred from the want of traditions about this bird, it is certain from more careful enquiry, that so little ignorant were the natives, as a matter of fact, that, so early as the first discoveries of 1839, they joined readily with the English settlers in their further search for them, at the same time making no mistakes or blunders about the objects they were looking for, as is distinctly affirmed by Sir George Grey in a letter to the Zoological Society, himself, from having mixed as much as any one with the people, being a very competent witness on this subject. Indeed, in a subsequent letter to Mr. Mantell, Sir George Grey states definitely that when he first came to New Zealand, the natives told him that the Moas were fast disappearing, but that they thought one might, perhaps, be found—and Mr. Mantell has, incidentally, pointed out that the corroded state in which the egg-shells of the Moas are often found is no test of their antiquity, nor caused, as was at first supposed in Europe, by the long continued action of water; it is rather due to the wearing influence of drift sand, especially at Waingongoro where Bishop Williams procured his first specimen, a place, by the way, which the natives, then resident there, asserted was the spot where their ancestors had first landed. Nor are we, indeed, now without direct testimony on this head; for the

natives asserted to Governor Weld that the Moa, like the Emu, defended itself by trampling on it adversary, and warned him not to go behind them as they kicked like horses;*Hector, l. c. two facts, which it is scarcely possible to suppose were purely inventions of their imaginations. Again, we are told by Mr. Hamilton that he spoke with an old Maori in 1844, who remembered Captain Cook, and who said he had seen the last Moa, describing it as having a long neck like a horse. Mr. Pollack, too, in an early account of New Zealand, affirms the same thing, and states that the natives told him that when food was scarce, the birds were easily entrapped, an assertion then more probable, from the remark of the old man just mentioned, that the plan usually adopted for catching the Moas, was to drive a post into the ground before the caves they frequented, with a stout noose attached to it. Lastly, Dr. Hector himself noticed, in the neighbourhood of Jackson Bay, well-worn tracks through the high scrub about sixteen inches wide, and such, too, as could not have been made by any animal or bird now existing in the Southern part of the Island. It has also been stated, I believe on good authority, that dogs have been known to suck the Moa bones, shewing clearly that these specimens, as would also have been the case with the skin and and muscles of the neek recently found, must have retained in them some nutritious matter. Putting, then, all these statements together, I confess I do not see how any conclusion can be arrived at, but that the final extinction of the Moas is quite recent. Professor Owen has, I believe, supposed that the Dodo and the Moa passed away together, probably about two centuries ago: but there seem now, fair grounds for thinking, that some specimens of the latter were really alive, at least in the most Southern parts of the Middle Island, as lately as the commencement of the present century.† Hutton, “Trans., N. Z. Inst.,” Vol. VII., p. 138. It has been stated by Dr. Hector, that the character of the plains and of the brushwood in the recesses of the province of Otago, are pecularly favourable to its habits. In concluding, then, this portion of my paper, I think I am entitled to say, that, so far as the story of the Moa goes, the credit of the natives, as a truthful race, is unimpaired; and that their not having preserved in their traditions any special references to it, nowise affects the truth of their further assertions of having first colonized the Northern Island about 500 years ago.‡ Vide post, Art. II.—[Ed.] (2.) Ethnological. I take next, the connexion and affinities real or imaginary, between the

Maoris and other peoples, as inferred by European Scholars, on Ethnological principles or from peculiar existing customs: in other words, I propose to examine the questions “Are there any populations in the Pacific Islands with whom there is reason to suppose that the Maoris have blood relations, either by parallel descent from a common ancestor or by a more immediate and traceable pedigree, or do they stand alone and with no apparent affinities with any one else? As my object is, chiefly to place before those, who may look into the Transactions of the N. Z. Institute, the evidence about the Maoris which seems to me reliable, I have not thought it necessary to quote at length the various views that have been held on this subject by scholars in Europe, such for instance, as W. von Humboldt, Crawfurd and others. I have thought that it would be more useful to consider, chiefly, the theories of those writers who, like Mr. Thompson, have placed their opinions on record in the pages of this work.* I have made one exception in the case of Baschmann, “Iles Marquises,” Berl., S. 1843, to whom I have referred constantly in the latter portion of this paper for views of the Analogies between the Talitian and Hawaian dialects. Now before I proceed to make such observations as seem necessary on this part of my general subject, it is necessary that I should state that ethnologists are, generally, agreed in dividing the inhabitants of the Pacific Islands (meaning by this term those portions of land which lie between the two great continents of Asia and America), into certain leading groups; though I must add, in my judgment, not unfrequently, with very inadequate designations. On the whole, I think the division into five such groups, now usually accepted on the continent of Europe, is the clearest and best, though not wholly free from objections: I shall, therefore, adopt this here; though it is not an exhaustive division and many instances occur, as might naturally be expected, of the overlapping of the lines of partition, and of the intermingling of distinct but adjoining populations. The names of these divisions are: I. Malaisia. Comprehending Sumatra, Java, Borneo, Celebes, Moluccas, Sooloo, Philippine Islands, and a considerable part of Malacca and Formosa. The characteristic of this people is that they have brownish-yellow skins and lank black hair. II. Melanesia. Comprising New Guinea, Arra, Mysol, Waygeon, New Britain, New Ireland, New Caledonia, New Hebrides and the Solomon Islands. These people are dark (nearly black) as regards their skins, with woolly and frizzly hair. The western

portion of this population is often known by the name of Papuans, and their abode as Papuanesia. III. Australia. With dark-skinned but rarely black population, with hair, however, not frizzly but lank and soft. IV. Micronesia. Consisting of several small groups of islands, many of them coral reefs enclosing lagunes, as the Ladrone or Marianne Islands, the Ludack Chain, Kingsmill, etc. The people who inhabit them are much mixed, and, in many of the islands nearly connected with the Melanesians. V. Polynesia. Comprising by far the most numerous groups, and extending from the Navigators' Islands on the west, to Easter Island in the extreme east, with those of New Zealand, the Friendly, the Society, the Austral, Hervey and Gambier groups, the Low or Saumatoa, and the Sandwich Islands. As distinguished from Melanesia or Micronesia, the inhabitants of these Islands have more resemblance to those of Malaisia, with light or dusky brown skins, often with a tinge of yellow, the New Zealanders and Sandwich Islanders (or Hawaiians) being the darkest, with black and curly as distinguished from woolly or frizzly hair. One other considerable group I have omitted, purposely, that of the Fiji (or Viti) people, as it would seem they are a very mixed race, with many affinities to Melanesia, though their grammar is more like the Tongan. I have not been able, at present, to meet with any very satisfactory account of them, but, as they have now placed themselves under the sovereignty of England, we shall soon I presume know whatever there is to be known about them. It is clear that a great many Polynesian words are incorporated in the few specimens of their language I have met with; indeed, in the name of their principal island Vanua-leva (the high land) I recognize at once, the Maori whenua. One marked distinction between the Viti and the dialects of the adjacent islands, which I have noticed, is the common occurence of two consonants at the beginning of their words, without an intervening vowel. The Fiji chieftains are said, too, to be devoted to the adornment and dressing of their hair, and to exhibit on their heads the circular mop-like masses of hair, so characteristic of the Papuans. It is probable, that the darker hue noticed as I understand chiefly in the case of the Maoris who are now found in the Southern Island is mainly due to their out of door life, exposure to the weather, and laborious occupations, The climate of New Zealand, especially southward of the middle of the North Island, does not differ very much from that of the South of England and France; requires, therefore, warm clothing and gives ample scope for bodily

energies;but the reverse of this must be the case with Saumatoa, Marquesas, Society and Navigators' Islands; while the climate of the Sandwich Islands must approach much more nearly to that of the latter, than to that of New Zealand. So far, then, I think it may be taken as a matter of general agreement that the native race of New Zealand have, on the whole, much more marked resemblance to those populations grouped under the name of Polynesia, than they have to the Negrito or Papuan peoples. It remains, however, to be shewn whether they have any nearer connexion with these Islands than may be fairly assumed from the broad differences between the black Negro and the yellow Malay; and, further, whether, admitting this one physical resemblance between them and the Malays, there is any reason to suppose them the descendants of a Malay colony, who might have found their way to New Zealand six or seven centuries ago. It is right that I should state; in limine, that some writers, as Dr. Dieffenbach, and, to some extent, Mr. Thomson* Mr. Colenso, I observe, asserts that the Maoris are not Polynesians, but I do not see that he has given any very strong reasons for this opinion. also, have maintained that there are two distinct races in New Zealand, the yellow-brown and by far the most numerous, but, besides these, a much darker skinned people; and have assumed that these two classes are descendants of two original stocks, the darker being the original. Later examination has not, however, as it seems to me, confirmed this view; besides it is scarcely probable that had there been, at any time, a considerable infusion of a Negrito population, they would not have left behind them some other traces of their former presence, than merely a certain number of darker skinned people, with hair differing in quality from that found among the majority of the population. I shall recur to this theory, presently. Now there are various ways in which such an inquiry as I am proposing might be carried out, independently of what are called “Race-characteristics,” such as the recognition of a similarity or peculiarity of customs, manners, etc., prevailing through all or most of the leading Polynesian groups, but which are found less universally, or not at all, elsewhere;—or, the unity or difference of dialects among the islanders. To each of these, especially the latter, I shall refer at some length hereafter. But I must notice first, a new view of the “whence” of the Maoris, which has been advocated with much ability in the Fourth Volume of the “Trans. of the N. Z. Inst.” by Mr. J. T. Thomson; because, if his theory can be maintained, the Maori can, hardly, be any longer considered as a

leading member of the Polynesian groups; at least, I do not understand that Mr. Thomson applies his theory to the whole of the Pacific Islanders, or perhaps, I am more accurate in saying, to the same extent, as in the New Zealanders. In this theory, Mr. Thomson, following, often in the same words, the very learned, but, to my mind, unsatisfactory views, of the late Mr. Logan, in the “Journal of the Indian Archipelago”* The elaborate papers of Mr. Logan on this subject are in the fourth, fifth, and sixth volumes of that work (for 1850, 1, 2). I do not see that, except in his researches in the Appendix to “Trans. N. Z. Inst.,” Vol., VI., Mr. Thomson has added much to what Mr. Logan gave to the world, 20 years ago—soon after which time, I remember reading them—while, he has, in many instances, adopted the exact words and phrases of Mr. Logan. Coming as they did from a man of such linguistic eminence as Mr. Logan, they, naturally, attracted much attention; it was, however, very generally thought, that his data were not sufficient for the very wide generalization he deduced from them. has assumed, that, in remote ages, a much wider range of country was occupied by the dark skinned and woolly haired races; in fact that they ranged over the whole of the plains of Hindostan, as well as over Africa, Madagascar, the Andaman Islands, New Holland, and New Guinea, etc., indeed, as far as 170° E. long. Closely, on the northern flank of these dark men, were the energetic Aryans, who, at some time or other, forced their way so far west and north-west as Ireland and Scandinavia, and the Tibetans—that is, the White and theYellow races—both of whom, ultimately, though, probably, with an interval of many centuries between them, descended into India, the one by the Punjaub, Jumna, and Ganges, the other by the Brahmaputra. The result of these invasions was (though chiefly through the agency of the Tibetans, for the Aryans have never much influenced Southern India), the expulsion or, more probably, the enslavement of the dark races, so far, at least, as India was concerned. I may add that it was a further view of Mr. Logan, that some of the castes in the South of India shew in their physiognomies a strongly marked African character, a proof, to his mind, that they are remnants of an Archaic formation of a still more decidedly African type. Thus, he says, the black Doms of Kamdon have hair much resembling wool. But how, one naturally asks, did they get to India? So far as we know, the genuine Negro of Africa has never been a navigating race: and the same thing may, I believe, be predicated of the Papuans and of most of the other Negritos; and though there may have been conquerors from India, who, reversing the fables of Sesostris and Semiramis, may have brought from Africa an entire slave population and settled them in India; as history is

wholly silent on this point, I cannot go the length of accepting, still less of proposing, so wild an hypothesis. Not the least objection I have always felt to most of Mr. Logan's theories—as distinguished from his extraordinary linguistic knowledge, is the prodigious length of time required for working them out. It is true, that if the Logan-Thomson views could be proved, some of the difficulties of the “whence” of the Negrito Island races would be got rid of—for, in such a case, one might suppose the “Melanesian” occupation of Timor, Gilolo, etc., due to their expulsion by the Yellow man from India; and, further, that the Yellow tribes, now, generically, called Malays, may be descendents of those Tibetans (to call them so ex hypothesi) who, coming down from Central Asia into India, drove the dark skinned people before them. I am not called on here to discuss this question; nor, indeed, would it be possible to do so, within the limits of any one paper; but it is worthy of remark, that, though, in some of the islands, the wild dwellers in the inmost fastnesses are as fair as the Malay coast-men, in other islands the dark people have been evidently forced back into the interior, while the yellow races have secured the sea shores and, with these, all the trade of the neighbourhood. Hence, there is no improbability in the idea, that the Yellow men did effect certain conquests over the Negroes, though it does not follow that India itself was ever populated by a purely Negro race. * I may as well notice here, that the presumed Negro occupation of India could not have been called the “Barata Kingdom” (more correctly Bharata) as Mr. Thomson at least implies, in his subsequent paper on the “Barata Numerals” (“Trans. N. Z. Inst.,” Vol. V.) Bharata-varsha. (“Bharata Kingdom”) is a title essentially and purely Sanskrit, and could not have been applied to any Negro dominion. Bharata was the son of Dushyanta, and India was, hence, called his kingdom. With regard to the Maoris, Mr. Thomson thinks (as judged by their features) that they are “clearly a cross,” with affinities to the Dravidian or oldest inhabitants of the South of India. But this view is, obviously, at variance with the Negro theory—for the Dravidians are certainly descendents of a Yellow race, † Professor Max MuUller long since in “Bunsen's Philosophy of Universal History,” Vol. I., London, 1854, demonstrated the close connexion between the Dravidian or Nishada races and the so-called Turanian population of Central Asia; and his views have been completely confirmed by Dr. Caldwell's admirable Dravidian Grammar (2nd Edition, 1875). Though possibly connected with the Finns, the Lapps, and Samoiedes of the North or with the Basques of the South, they have, assuredly, no Negro affinities. Nor, without much more information about the “Tamil books” that Logan has referred to, should I venture to conclude, that the people with “tufted hair” said to be mentioned in them, were what we understand by “Negroes. who, according to it, drove the Negro people out of

the country; and have, except accidentally, nothing in common with the Negroes. No doubt some of the Tamil population are dark enough, much darker than the ordinary Malay; but their hair is as a rule of a soft glossy black, the very opposite to that of the crisp and woolly Negro. Far more probable, is a further suggestion of Mr. Thomson, that the Maoris are, in part, offsprings of the Tibetan and Ultra-Gangetic races—which Mr. Logan has, I think, also proposed, a race, perhaps, now represented by the Bajow or Oranglaut;—(“Men of the Sea”) the more so, that these tribes are, in an especial manner, “Sea-nomads” and frequent to this day all the waters and islands of the Indian Archipelago. In this way, no doubt, it will be quite possible for New Zealand to have been peopled—only, that unless this took place, at a very remote period we should unquestionably find much more modern Malay in the Maori language, than either Mr. Crawfurd or Mr. Thomson have been able to point out. * I should add that Mr. Thomson has given a very interesting account of the people whom he has met with in India—illustrated by his own sketches—with certain inferences from their physiognomies. With these views I do not presume to interfere—but I may be allowed to remark, that, with the exception of some very decidedly marked varieties, such as the Negro as compared with the pure Hindu or the pure Caucasian, individual examples from drawings or even photographs are not perfectly satisfactory. We want the presence of “numbers” before our conclusions can be safe. So in language—the occurrence of a good many individual words—without grammar or syntax—is nothing worth, as an evidence of the origin of the people among whom they may happen to be found. Moreover this theory does not account for the supposed “cross” unless we imagine the invading Bajows to have brought with them a handsome supply of Papuan slave girls. With Mr. Thomson's further dictum, that the obliteration of an intervening race does not destroy the Ethnological links between two distant regions, I should, of course, agree—only that I do not perceive, in this case, any need for such an obliteration: he has not, I think, shewn that the actually occurring cases of this “cross” are very numerous; while, so far as I can learn from other sources of information, it would appear to be generally considered that the Maoris are one in race as well as in language. Mr. Thomson in his next paper “On Barata Numerals” (“Trans. N. Z. Inst.,” Vol. V.) endeavours to support his view of “Barata expansion” by an elaborate comparison of the numerals of 34 islands and districts with those now in use in New Zealand, drawn up with great care from the works of Logan, Earle, Wallace, and others, and maintains that the remarkable similarity he has, in many cases, succeeded in showing, is due to these places having all, at some time or other, been either colonized directly, or greatly influenced by the so-called “Bhàrata” population. Now, as I have said

before, I am not at all convinced by the mass of erudition in Mr. Logan's papers, which I have gone through more than once with as much care as possible; moreover I believe, that present physicists are, by no means satisfied with reference to any near connection between the Negritos of the Islands and the genuine Negroes of Africa, though their external resemblances are, at first sight, considerable. I feel therefore, inclined to suggest this further query—viz., Is it not quite as probable, on the whole, that what Mr. Thomson calls “Bhàrata numerals” are really those worked out, gradually, by the colonizing Yellow man, and that, if they are now found, also, in regions occupied by the dark races, they have been forced upon the latter by the power, or possibly, by the intelligence of the former? At any rate, this hypothesis does not require the extreme length of time demanded for the Logan-Thomson theory. Nor, indeed, is it without some confirmation from what may be seen in Mr. Thomson's own list of numerals (“Trans. N. Z. Inst.,” Vol. V., p. 137)—for I observe, that, in New Caledonia and Arru, there are only two numerals the same as those of the Maori; in Kissa and Tenenbar, three; in Mallicolo and Tanna, and Vialo (Temir), four; all of these islands being essentially parts of Melanesia. Supposing, therefore, no sufficient evidence adduced to the contrary, the presumption would, I think, be that these Negritos had acquired such of their numerals, as are similar to those in Polynesia, either by compulsion on the part of their Yellow neighbours or conquerors, or, in the course of commercial intercourse. No one would, I think, assume from such proportions, that the Maori numerals, and those of the other islands, believed to be in many ways cognate with New Zealand, were derived from the ancestors of these Melanesians. Again, if Mr. Thomson's “Bhàrata” theory were true, we should naturally expect some resemblance—and a near one too—between these “Bhàrata” numerals and those of the languages connected more or less nearly with Tamil, the present representative language of the South of India—the Dravidian Tamils—having, agreeably with Mr. Thomson's view,—succeeded to the territorial possessions of the Negro race, they expelled—or, rather, being the actual Dravidian race (according to another of his suppositions) who have led to the “cross” he notices among the Maoris. All I can say, on this head, is that I have carefully examined the numerals in all the leading languages of Southern India, the Tamil, Canarese, Carnatiku, Telagu, etc.—and, that I have not been able to detect even a solitary resemblance with any of those in Mr. Thomson's list. If then, similarity of numerals be any real test of the connexion of races (which I do not at all assert to be the case) it is clear that Mr. Thomson's argument for the existing numerals cannot be sustained as any evidence of the “whence” of the Maoris. If

the present Dravidians were nearly connected with the Maoris, their common numbers would have been nearly the same—if not identical. * After all, I venture to doubt how far numbers are a safe test of race. The resemblance of numerical systems ascending to high numbers may be, as demanding considerable power of abstraction, but the simplest and smallest numbers up to 5, would seem to be within the reach of the most unlettered savage. But though I have thus freely criticised the views put forward by Mr. Thomson, and cannot admit I am a bit more convinced by his reasonings than I was, years since, by the still more elaborate papers of Mr. Logan, I recognise with pleasure the patient labour and study he has shewn in the papers he has contributed to this work, and the value of his independent researches in connection with Malagasi, Malay, Tongan, and Maori. I further think, that it would be a most valuable work, if any scholars, who have the time and the means, would subject the languages of Africa to the same exhaustive treatment, that has been applied with such remarkable success to the languages of Europe, by Gormin and to the Sanskritic dialects by Bopp. Were such a work to be effectually carried out, and were the result this, that the numerals of any reasonable number of these African languages or dialects were found to agree with those in Mr. Thomson's list, I would be first to recognise this fact, and to withdraw the objections I, at present urge. But I must confess I am not very hopeful of the proof of any such agreement between the numerals or, indeed, with any other linguistic system in Africa or in “Indonesia”—the more so, that a very intelligent Negro,—himself a native of the West Coast of Africa, and at present a student at Oxford, tells me that, though familiar with four or five languages on the West, he cannot understand one word of the Eastern language of Zanzibar. On the other hand, I quite agree with Mr. Thomson, as to the principle of investigation to be pursued in tracing out cognate languages—and the primary words (as he aptly calls them, the “fossils” of language), in that they express the first wants of man, are more tenacious of existence than any others. No doubt, common nouns, pronouns and verbs, when found little changed in a long series of dialects, do go far to prove descent from some one common source. It is, on this very principle, that we speak, and speak truly, of the Celtic, Slavonic and Teutonic languages being akin with Sanskrit, not, indeed, as children to a mother, but as brothers and sisters, the offspring of a parent, at present nameless. As the Roman poet, said so long ago, they may be termed sisters with a strong family resemblance,— “Facies non omnibus una nec diversa tamen, qualem decet esse sororum.” But though, as I have said, I cannot accept Mr. Thomson's theory for the

derivation of the Maoris, except so far as he has shewn in his able comparison of the Malagasi, Malay and Tongan languages, I think that there are certain customs prevalent among all or most of the Polynesian Islands, which peculiar to or characteristic of them, do tend decidedly to shew that they were once one people. Thus there is, or has been, in most of them, the worship of a Supreme Being, Tangaloa generally, in New Zealand, Maui, and what is remarkable an almost total absence of Temples, or of anything but the rudest form of Idols. At Tahiti, indeed, and in the Sandwich Islands, coarse wooden figures, which served for such, are noticed in the early missionary narratives. In the Ashmolean Museum at Oxford there is one in stone said to have been brought from Raiatea; and the curious Colossal figures from Easter Island, two of which have lately been set up in the British Museum, may, very probably, have been worshipped by the original population, a different race from the present inhabitants, and perhaps, as has been supposed, of Mexican origin. There seems, also, to have been in Tahiti certain sacred precincts not unlike the old Greek Temenos. * There are several words in Hawaian for image as Tü (in the Dictionary Kü) as Kü akua (Tü atua) He Kü, etc. The last is, probably the same as the Maori Heitiki—a charm worn round the neck. In the sacred precincts, animals constantly, and human beings, occasionally, were sacrificed, to please the good or to appease the evil Spirits. The majority of the Gods, however, were deified early chiefs: much as, in ancient Greece, the hero and the God were often nearly connected together. There is, also, much similiarity in the accounts given of the origin of the different islands; that of the Tongans, that they were fished up from the bottom of the sea being a likely story enough for people situated in a vast ocean like the Pacific. In several of the Islands, their Paradise is placed in the far west. Is this an indication of the traditional history of their emigration? Mr. Logan has also pointed out that the leading chiefs, generally bore a title, variously pronounced Aliki, ali'i or ari'i. † But to suppose as Mr. Logan further does (“Journ. Ind. Arch.,” Vol.IV., p. 355, and note) that Aliki, etc. is the same as the Indo-European “Aryans” seems to me something like Philology run mad. On such a principle any thing may be derived from any thing as poaka from the English “pig,” or “pork,” or “kuri” from “cur.” Thus in Maori, ariki or whaka-ariki means a man of high or ancient hereditary descent, and one, therefore, clothed with a peculiar sanctity. Mr. Mariner gives many interesting details of the Religion of the Tongans, before Christianity, and much similiar information may be gleaned from Ellis's Polynesian researches—such as the institution of what he calls the Areoi, a set of wandering players—who devoted themselves to every kind of debauchery—were hated by the agricultural people whom they plundered—but upheld by the chiefs—and, generally, looked up

to by the populace at large, as being something Divine, or rather Diabolical. This institution was directly connected with the very prevalent practice of infanticide, especially with the destruction of female children. It would seem that, by some fortunate accident, the customs incident to this institution were not accepted by the Marqesans. Human sacrifice was universal, and Suttee (Satî) not uncommon. Again, there is the institution (probably sanitary), of Tapu (anglice, Taboo), * There is no doubt that the Tapu was a wise provision (made a religious ceremony in order to enforce it more completely) for the purpose of preventing the spread of infection etc., particularly, leprosy. (See Hector, “Trans., N.Z. Inst.,” Vol. VI., p. 370). which, extending to all the Islands of Polynesia, is, in an eminent degree, characteristic of them—though not absolutely unknown elsewhere. Mr. Crawfurd used to say that the name and the practice were of Indian origin † It is true that there is a Hindu word of common use (tapas) in the sense of “penance,” “ascetic devotion,” “self-torture,” etc., the like—and that, from it, we get such derived forms, as Tapaswi, a devotee; Tapasga, austere devotion; Tapasani, a female devotee; Tapodhana, one rich in devotion—who leads a life of penance; and, in Guzerati, the same word is used for the servant and minister of a Temple—but, I confess I cannot see much connexion between these meanings and that universally given to the Polynesian Tapu.—I don't know what word, if there be one, corresponds with the Hindu Sati. But the burning the widow in honour of her déceased husband was not an universal custom in Polynesia, moreover, is, in India, an atrocity invented since the laws of Manu.—but this may, I think, well be doubted. It is not, however, easy from the very various accounts of it, to decide,—what may have been its most probable origin: its universality, however, proves its antiquity. The Hindu, tapu (penance) is far from being comprehensive enough; I should rather have supposed its origin in the will of some great potentate or chieftain, who united in himself, as was so often the case in former times, not only in Polynesia, but in classical lands, the double office of King and Priest. Obviously, no institution could tend more fully to foster and support the tyranny of the leading men. While the quite recent, if not still prevailing custom, of disusing certain words or syllables, which, occurring in the names of great chiefs, might be supposed disagreeable to their ears, is, I believe, but one further instance of the power of “tapu.” It must be remembered that the usual Maori word for a priest “Tohunga” does not necessarily, bear the title now assumed for it—but is strictly the “skilled artisan” ‡ I suspect the greater number of Tohungas belong to the only tribe which is skilled in wood-carving. They live near the East Cape, and are, frequently, yellow-skinned. Individuals of this tribe are sent for by the other Maoris all over the Islands to do wood-carving for their houses and halls. See “Trans. N.Z. Inst.,” Vol. I., p. 446.—J.H.—the clever fellow, who can turn his

hand to any thing. * The oldest English name is believed to be Pratt (the family name of Lord Camden). This is “praet,” the “ready” man. This is clearly seen in Tongan, where tufunga means any kind of workman as tufunga ta maka, a mason; tufunga fei cava, a barber. In an early stage of society, such men naturally take the place of leaders, and if they could add a little superstition to their other abilities, this would help them all the more to keep down the common people. Another custom very prevalent in the Polynesian Islands, though not strictly confined to them, was that of cannibalism. It is true that, in Australia, with a population quite as savage as can be found anywhere else, as also in Micronesia, man-eating was not practised,—moreover, that it was less frequent in the Navigator's Friendly Society and Sandwich islands than in New Zealand, the Marquesas, etc. Still, there can be no question that the practice was occasional everywhere and involved no loss of caste or character on the part of those addicted to it. In some places, too, it would seem that the victims, generally slaves or captives were fattened previously to being killed. Much has been written on this subject, and it has been held up to view as the most atrocious act that man can commit: it seems, however, to me to be, but one more instance of the entire disregard of anything sacred in the human body or in human life, which the stories in the works of Mr. Mariner and Mr. Ellis show to have been so generally prevalent in the Pacific Islands. It is not pleasant to call hard names, but there can be no doubt that, till very recently, murder daily and under every form, was the characteristic practice of all these Island populations. To eat portions of a body so slain—especially, if slain with the view of propitiating some evil spirit, is not unnatural, and has been done in modern times and by people calling themselves Christians. † So lately as in the insurrection of 1848, in the public streets of Palermo, and, during Garibaldi's war of independence, at Messina. A Sicilian Brigand just slain, is stated to have eaten the hearts of the people he murdered (Daily News, October 18, 1865.) Another custom, like most of those I have mentioned, very universal among these Islands, though, not absolutely restricted to them was that of Tattoo;—the carving on the outward surface of their bodies,—and, especially of their faces, certain patterns, generally curves, and forcing into the skin thus incised, various pigments most frequently of a blue colour. This custom, though partially practised by some of the Tribes now living in the Eástern outskirts of India, cannot, I believe, be traced to India itself. The word used for it is nearly the same in most of the dialects. Thus, Tahit., tatau; (with two special words given by Monkhouse and Cook for tattooing in different

parts of the body, viz., Tamoraho and Tamorau) Marqunesan, tatu and patu; Nukuhivan, pikipatu; Tongan, Tattu; Hawaian, Kakau; but, curiously enough, though as much practised there as anywhere else, there seems to be no similar word in the New Zealand language, except the doubtful Tamoko, or lizard, is the Maori word, possibly from the curved lines they rejoiced in tracing, in parallel lines, on both sides of the face. As to the origin of this curious practice, there is great diversity of opinion, some writers fancying that it arose from a sense of decency; but it seems hardly probable, that a people accustomed, in many of the islands, to wear scarcely any dress, should have adopted, for this reason only, a custom so extremly painful in its operation. * I do not know whether Polynesian skins are less sensitive to pain than those of Europeans; but if not, such tattooing as appears in the portrait given by Ellis of the N. Z. chief, Hongi, must have entailed years of suffering. Mr. Logan (I think) mentions a kind of tattoo, in some parts of Melanesia, which must be more hideous to look at, if not more painful in execution, than that of Polynesia. It consists in making great and permanent wales all over the body. A more likely reason would seem to be that of striking terror into their enemies; while, if it be true, as Mr. Ellis asserts, that the attendants of the different chiefs were usually tattooed like their masters, only less elaborately, this plan would answer well as a means of identification. Another curious custom is that of Cava-drinking, the nauseous mode of preparing which Cava, in the Tonga Islands, is minutely described by Mr. Mariner. I do not know whether this custom is universal, but the word is found in most of the dialects, for a species of the pepper plant. I think I have now said enough on the subject of some of the principal customs, which, if not all peculiar to Polynesia, certainly prevailed in these Islands more than anywhere else. They are, as it seems to me, essentially such, as would be handed down from family to family, and from tribe to tribe. They are, hardly, such as would be invented by two or three separate sets of peoples, but point, almost as surely as the colour of the skin or the texture of the hair, to a period, when the inhabitants of these widely scattered islands were one people dwelling together. I venture, therefore, to hope that in drawing to a conclusion, this, the Ethnological portion of my Essay, I shall be deemed to have shown some reasonable grounds for believing the Maoris, Tahitians and, generally, the dwellers in Polynesia, with the partial exception of the Fijeeans, One Race, physically, united under one group, with clear and definite lines of demarcation, which separate them from the Dark-skinned people on the one hand, and from the White races on the other.

(3).Philology, &c. In the Third portion of this paper, I propose to examine, so far as I can with the limited materials within my reach, the relation, if any between the Maori language as traceable during the last hundred years, and those of the other principal islands of Polynesia; to do, in short, on a somewhat more extended scale, what Mr. Thomson has so well done recently (“Trans. N.Z. Inst.,” Vol. VI., p. xxv. Appendix, 1873), in the careful comparison he has made between the Malagasi and Malay, Maori and Malay, Tongan and Malay. In this important paper, he has most successfully demonstrated, and, I believe for the first time with sufficient fulness, the connexion subsisting between these languages; an affinity, which has, indeed, been pointed out before by W. v. Humboldt, Buschmann, Chamisso and others, but, as resolutely denied by other competent scholars, such as the late Mr. Crawfurd. And I am the more induced to undertake this work, because, beyond what Mr. Thomson has accomplished, and a few remarks by Mr. Colenso in a paper not specially devoted to this subject (“Trans. N. Z. Inst.” Vol. I.), I do not perceive that this question has been taken up by anyone else in New Zealand, or, at least, has been discussed as fully as it deserves. I can, of course, only hope, in a very small degree, to supplement Mr. Thomson's researches, especially, as for one or two important branches of the whole subject, I have failed to procure either in London or in Oxford, the necessary books. * I had better state, here, the only books I have been able to see with reference to the present enquiry, are Kendal (or Lee's) “New Zealand Grammar, etc.” 1820; Williams' “Dictionary of New Zealand and Grammar,” 1852; Duff's new ed. 1874; “English Common Prayer in Maori;” Maunsell's “New Zealand Grammar,” 2nd edit., 1862; W. L. Williams' “First Lessons in Maori,” 1872;” Mariner's “Tonga Islands' Gram. and Dict.,” 1827; Andrews' “Dictionary of the Hawaian Language,” Honolulu, 1865; Buschmann's “Aperçu de la langue des Iles Marqueses,” with notes on the Tahitian language by Baron W. v. Humboldt, 1837; A. V. Chamisso's “Uber die Hawaischen Sprachen,” 4to, 1837; Mosblech's “Vocabulaire Oceanesia-Française,” 1843, with various papers published by J. R. Logan, in the “Journ. of Ind. Archæology.” I should have been most glad to have obtained more information about the Samoan, Low, and other dialects but could not. When I have referred to these, I must be supposed to be quoting (whether or not I specify the page) from the invaluable papers of J. A. Logan.—I have of course, had before me, many if not most of the Navigators of the South Seas, as Cook, Vancouver, D'Entrécasteaux, Dupont, D'Urville, with some other works on the subject, as Ellis' “Polynesian Researches,” etc., etc. Yet, I am in hopes, that this work, though, avowedly, so imperfect, may yet be so far useful, that it will place within a small compass, what seem to be the inost striking varieties, at least as regards their litteral system, between many of these dialects, and may thus enable

other students, on the spot, to follow up these matters, in greater detail,—and, as certainly, to correct many errors into which I have myself surely fallen, in my desire to call attention to those things, which seem to me of most interest and importance in the languages I have been able to examine. The general result, I am convinced will be, a complete and satisfactory proof to all who have time or patience to follow up the steps of the argument, that over a wide range of the Pacific Ocean, including the Sandwich Islands on the extreme north, and New Zealand and Madagascar to the south and south-west, are still to be found ample remains—the “disjecta membra"—of one original language. It may, perhaps, also tend to the solution of the ultimate and still more interesting question, whence, at a remote period, the forefathers of the present occupants of these islands, themselves, emigrated, I venture to add must have emigrated—for, in point of fact, Crawfurd's “autochthones” theory is far more difficult of comprehension. Given sufficient time, and, here, I have no evidence against me, even though I am not able to produce evidence in my favour which will convince other people; there is no difficulty, whatever, in conceiving a continued emigration from the East (if Mr. Ellis' theory be preferred), or from the West, which I hope, hereafter, to show is of the two the more probable. I think it highly probable that the researches of Mr. Thomson combined with the few matters I have, myself, been able to note down in the following pages, may, if more fully carried out by individual scholars at Rarotonga, Manganevu, etc., and, at perhaps, other less known islands, form a useful manual for future and more advanced study; or, at all events, a tolerably accurate record of our present knowledge of these islands, so far, at least, as their languages are concerned. It is, I think, a work that ought not to be delayed, as contact with European civilization—with its languages—together with the natural influence of trade, must every year modify, considerably, the native tongues. Mr. Logan (“Ind. Arch.,” Vol. IV., p. 272) says “I saw, lately, some Honolulu youths at Singapore for the first time. Their thoroughly English dress, manners, and speech, were calculated to make a strong impression, after a perusal of the account of Cook's reception and death at Hawaii in 1779.” I feel myself certain, that, not many centuries will elapse, ere Tahitian, Fiji (or Viti), and Maori, will be as much things of the past, as Cornish is now in England. With the extinction of these languages, it is not too much to say, that though “race characteristics may best go down in blood” (Whitney, “Life and Growth of Language,”) we shall lose an invaluable aid in our endeavours to trace out the “whence” of the Maoris.

I propose, therefore, now, as I believe this is, on the whole, the simplest plan, to take the different parts of speech, separately, in the order usually presented in European grammars, and, in this instance, to adhere to that given by Dr. Williams, as, perhaps, the clearest, though, by no means, the fullest. Dr. Maunsell's work is invaluable for the Syntax; but, with this portion, I do not feel myself at present competent to deal; nor is it, indeed, essential that I should for the object I have here in view. I ought, perhaps, to add here, that Dr. Maunsell and some other writers have objected to the use of the names, cases, moods, etc., in Maori grammar, because, in it, strictly speaking, as in other Polynesian dialects, we do not find those modifications of the root-stems of the words which occur in Sanskrit, Greek, etc., and which constitute the peculiarity of what are, therefore called Inflectional languages. I venture, however, to think that these names may be usefully retained as conveying, at least, to European eyes and ears, certain definite senses which are in the main true. As a matter of fact, such names might as easily be dispensed with in most modern languages, and are, indeed, only kept for the sake of convenience. Before, however, I proceed to discuss in detail the various parts of the Grammar, it is necessary to note the principal and constant variations in the consonantal systems of these Oceanic languages and to establish for them so far as it is possible, the common laws of their permutation, after the fashion so successfully applied by Jacob Grimm in the case of the European tongues. There is, indeed, a remarkable regularity in these changes; but I am not, at present, prepared to state that the reason for this is, that any one of them stands to any other exactly in the relation of parent to children. This point I must reserve till the conclusion of my paper. Now many writers have called attention to a larger or smaller number of these variations. Thus, Mr. Thomson, in the article to which I have already referred more than once (“Trans. N. Z. Inst.,” Vol. VI., p. 53), has pointed out that, in Tongan, the Maori p, t, k, r, and w, find equivalents in b, b, g, l, and v: as, in the following instances,— Maori Potiki Tongan Bibigi (child). kuri guli (dog). wera vela (hot). so, too, h and p similarly interchange, as, Maori hua Tongan pua (fruit). huru pula (hair). Again, Mr. Logan has shewn that k is very generally omitted in Samoan, Tahitian, and Hawaian, though present in other dialects, and, in the same way, l in Tongan; that h appears as s, in the dialects of the Samoans and

Fakaafo; f as w or h in Maori, and h in Hawaian; that the v of the others, is generally w, in Maori, Hawaian, and Saumatoan, and that the Maori r, is replaced by l, in the Fakaafo, Samoan, and Hawaian. Other changes I have noticed (and I doubt not more are to be found by learned scholars in the islands themselves), such as the double one, in the case of the Maori wahine (woman) as compared with the Tongan (fafine); the absence in Tahitian of the Maori ng; * Two-thirds of the Maoris use k for ng.—J. H. and, further, the change of the ng into n in Hawaian, and into k in Nukuhivan. The general inference to be drawn from these modifications would seem to be, that the Polynesian dialects (at least so far as we can judge of them from their present forms) do not sharply distinguish between v and w, d and t, l and r, d and r, b and p, g and k; that t and k are sometimes confounded; and, that one island has an affection for one set of sounds, another for another. In these changes and modifications of sound, the classical scholar is reminded, at every step, of the dialectical changes of ancient Greek, dependent as both the Classical and Oceanic dialects alike have been on the greater or the less education of the ear in different localities. † It is not easy to find words sufficiently distinct to avoid tedious repetition, or the ever-recurring tendency to fancy similar words in one dialect are derived, the one from the other. It would be tedious to say invariably “found in” or “occur in,” instead of calling what we observe simply a change. According to my view, it would be more accurate to say (for instance) that the Maori word wahine takes the form in Tongan of fafine; rather than to say that w and h respectively are changed into f: all, however, I mean to urge is, that if, for argument's sake, a single original Polynesian language be imagined, then, Maori, as a rule adopts one set of consonants, Hawaian or Tahitian another. I should add that Mr. Thomson gives the following proportion of the number of consonants in different dialects to which he has called attention; and this list is certainly curious, as shewing an apparent diminution in the powers of vocal expression, as you proceed from West to East. Thus he states that, while Malay has 18 consonants, Mindanao has 16; Wagi of Celebes, 15; Tanna, 13; Malagasi, 12; Mallicolo, 12; Awaiya of Ceram, 10; Tahiti, 9; Maori, 8; Marquesas, 7; Sandwich, 6: but it should be remembered, that some of these, as Mindanao, Mallicolo, and Ceram, do not fall within what is usually termed Polynesia; moreover, I am not satisfied (not that I doubt that Mr. Thomson has taken his lists from books correctly), that the numbers given above do really represent all the consonantal sounds—which accurate ears, combined with sound philological knowledge, would detect in even the existing languages. Mr. J. E. Alexander, who has written an excellent preface to Mr. Lorrin Andrews' valuable Dictionary of the Hawaian language, remarks

as general laws prevailing throughout the Polynesian dialects, that every word and syllable ought to end in a vowel; that most of the radical words are of two syllables; that the accent is usually on the Penultimate; and that the islanders have, as a rule, sharp ears for the distinguishing of vowels, but dull ones for consonants. He, also, points out as characteristic of Hawaian, and of one or two other of the dialects, that where the k (found elsewhere) is omitted, there is, in the middle of the word “a peculiar guttural catch or break.” It would be a very interesting subject for research, if missionaries or others, dwelling in the separate islands, and well acquainted therefore, with their marked geographical features, would notice whether any (and what) differences exist between the dialects of the mountaineers and of the low country people, and what are the laws governing the differing pronunciations of languages, presumably the same radically. Supposing, for instance, there was still a numerous race of natives living in the Canterbury Plains, these people would almost certainly speak in Maori, dialectically diverse in many ways, from the Maori of the North Island, or even of the west of the Alpine ranges of the Middle Island. Languages in narrow valleys between lofty mountains are often strangely different the one from the other, though, unquestionably, those of one family. As it is of great importance to keep clearly in view the nature of these changes, I submit, here, a list of them from the Preface to the Hawaian Dictionary, drawn up I presume by Mr. Alexander, as it very clearly shows what we may expect to find in each case. This list differs, in some slight matters, from the opinions quoted by me from Mr. Logan, but, having been drawn up on the spot, may, I have no doubt, be quite depended on. Fakaafo. Samoan. Tongan. Maori. Rarot. Tahitian. Hawaian. Marquesan. F F F W or H Wanting F or H H F or H K * The apostrophe is to shew the omission of the K mentioned before as producing “a guttural catch or break.” K K K ' ' K L L L R R R L Wanting M M M M M M M M N N N N N N N N NG NG NG NG NG Dropped N NG, N, or K P P P or B P P P P P S S H H Wanting H H H T T T T T T T or K T V V V W V V W V The meaning of this is, that when a word in one dialect begins with a certain letter, it will probably be found (if, indeed, it occurs at all, which is by

no means always the case) under the corresponding letter in the preceding list. The following list of words, taken from the same preface, illustrates this matter even more clearly,— Fakaafo. Samoan. Tongan. Maori. Rarot. Tahitian. Hawaian. Marquesan. Foe Foe Foe Hoe Oe Hoa Hoe Hoe Tonga Tonga Tonga Tonga Tonga Toa Kona Tonga Tona Sina Sina Hina Hina Ina Hina Hina Hina Ika I'a Ika Ika Ika I'a I'a Ika Vaka Va'a Vaka Waka Vaka Va'a Wa'a Vaka Songi Songi Hongi Hongi Ongi Hoi Honi Hongi Tufunga Tufunga Tufunga Tohunga Taunga Tahua Kahuna Tuhuna Kupenga ‘Upenga Kupenga Kupenga Kupenga ‘Upe’a Upena Kupeka In the same preface, it is further remarked that r and d are often hardly distinguishable from one another, a fact, which I notice has been preserved in Lee's (or rather Kendal's) “New Zealand Grammar,” where dua, todu, dima, and wadu are found instead of the present spelling of, rua, toru, rima, and waru; that, in writing Hawai, k has been “erroneously” * If this adoption of the k for the t be really an universal error, as implied in this statement, it cannot be too soon corrected, as it may gravely mislead those students, especially in England, who are attempting to trace the inter-connexion of the different dialects of Polynesia. But this must be done, if at all, at Honolulu, and by authority. adopted in the place of t; and that, throughout the dialects, there are comparatively few changes of vowels, and, when these occur, they are usually owing to consonantal influences. Thus, in Hawaian, hohu, honua, maia', and maika'i represent the feta, fenua, and maitaki of other dialects. Mr. Alexander, also, thinks that, in the consonantal sounds, the Hawaian is one of the softest and most attenuated of the dialects, being surpassed in that respect only by the effeminate Marquesan. I proceed now to take the different parts of the Grammar in the order set forth in Dr. Williams' “Grammar,” noticing— I. The Articles; and, here, I at once observe a very general agreement, such modifications as there are, applying usually only to the initial letters. Generally, with the exception of the Tongan, each dialect has a Definite and Indefinite Article. Thus Maori has te and he; Tahitian, Raro-tongan, and Mangarevan, te and e; Hawaian, ke (for te) and, sometimes, it would seem from Buschmann, he; in Tongan, he is the only article, but ko is often used in answering a question, as, koa tangata, a man. When several things come together, ko is generally used and he omitted, and so before proper names—as, ko Tuo, ko Koumete, ko Finow. Dr. Maunsell has further, I think, rightly considered that the indicator of the Plural in Maori (nga)

partakes of the nature of an article as do, also, te tahi (literally the one) and its plural etahi. * The forms corresponding with etahi in the different diálects have a strong family resemblance. Thus— Rarotongoan has tetai, etai. Mangarevan mai. Hawaian tahi and tetahi (and wahi). which is also found in New Zealand. Samoan sa, setasi, letasi, etasi. The Tongan (like the Mangarevan) is different—viz., na, foenihi (compare here the Samoan, nisi), and etaha.The latter, he thinks, corresponds very nearly with the use of the French des, or the adjectival “some” of English. Thus— Te tahi maripi, a knife; maku etahi ika, give me some fish. He shews, also, that te is used 1. Where no article is required in English, as, he kino te tutu, disobedience is sinful. 2. In the place of the English a, as, he mea kaha te hoiho, a horse is a strong thing. 3. Sometimes for the pronoun “some,” as, kei tahaetia e te tangata, lest it should be stolen by some one. 4. Before proper names, as, Te Puriri, etc. Somewhat similar variations may be noticed in the case of he, which, like te, is often used where no article is required in English, in the sense of “some,” and, before nouns in the plural number, as, kawea he wai, fetch some water;he tini oku kainga, my farms are many.† Buschmann (p.168) points out that, in Tahitian, taoua….ra sometimes occurs for the Definite Article; and te hoe, maa, te maa, and te hoe maa for the Indefinite. I must mention here that Mr. Logan, to whom all students of “Oceanic” Philology are so much indebted, has pointed out ‡ “Journ. Ind. Arch.,” Vol. VI., 1852. the curious fact that, in what he calls W. Indonesia, i. e., Sumatra, Java, etc., the definite article is si: this si, he thinks, is nearly connected with the Polynesian tahi, tasi,—and, probably, with the modified forms se and he. In like manner, other forms such as iti, ti and te, resemble closely the Polynesian te and ta. Again, an and ang, which occur frequently in an appellative sense, have a striking similarity to the Polynesian na, nga; while the ka, kua, and koe of Maori, Tongan and Saumatoan, would seem to be connected with similar forms in the dialects of W. Indonesia. In a former essay, § “Journ. Ind. Arch.,” Vol. IV., 1850. the same distinguished scholar has shown, that there are many characteristic features in Polynesian, which have not been preserved, either in Sanskrit or in any of its modern derived dialects, but, which are, at the same time,

found in Greek and other western Indo-European languages. Thus, this very definite article, lost in Sanskrit or Malay, but common in Greek, has remained in full use in Polynesia and, what is more curious, in even the mountain dialects of the rude Khasias of Assam. II.Substantives. Are distinguished according to their gender, number, whether singular or plural, and case. Of these, the first, in most of the dialects, is shown by attaching to the word, another one signifying male or female. Thus Maoris use tane and wahine when applied to human beings, tourawhi and uwha when applied to the brute creation or inanimate objects—as, He matua tane—He matua wahine—He kararehe tourawhi (a male beast)—He kararehe uwha (a female beast). Maori has, also, according to Dr. Maunsell (p. 19), several distinct words for specially related men and women as Tuahine, a man's sister—to which it is not necessary to refer to more fully here. The words and their use in the other dialects are nearly the same. Thus:— In Tongan and Samoan, tane, fafine. " Rarotongan, tane, vaine. " Tahitian, tane, vaine. and for animals, oni and ufa (Buschm., 168.) " Hawaiian Kane, vahine. The Plural is shown in several ways, but most simply in Maori and Rarotongan by the prefix of nga, as, tangata, a man, nga tangata, men. There are, however, in Maori other methods of expressing plurality such as placing before the noun some of the plural or dual pronouns, as, aku tupuna, my forefathers. * The singular may also be denoted by a singular pronoun as toku paraikete—my blanket. Sometimes o is used as, kei o Hone matua, with John's uncles; sometimes the ground form is altered—as, tamaiti, a son, tamariki, children. Occasionally ma is added, with, as Dr. Maunsell suggests, the Greek sense ὂὶἀμει,ὂὶπεϚι (viz., a person and his company)—as, kei a Kukutai ma, with Kukutai and his party; while a constant repetition of the same act may be designated by a reduplication of one or more syllables, as, kimo, to wink, kimokimo, to wink frequently. † When a special stress is needed to show that only one object is meant, Tongan inserts the particle be (only), as, togi be taha, axe only one.

In Tongan the plural, in the case of things inanimate, is mostly denoted by the particle e, combined with a numeral, as, togi e ua, axes two; falle e tolu, houses three, etc., or, when an indefinite number is required, by lahi, many or several; as, lahi e vaka, many canoes.† In animate objects, a distinction is made, as to whether they are rational or irrational beings, the particle toko being used with a numeral, in the former case; as, for instance, tangata te tokotahi, one man only; fafine toko toru, three women; tokotahi e tangata, many men. Two other words (probably old collective nouns) sometimes occur, viz., kau and tunga, as, kau or tunga tangata, a body of men; kau tangata tokoterau, a body of men, a hundred. In Hawaiian, the Plural is shown by na, pue, mau, with article te, hui, feia according to Buschmann, and, in Marquesan, by mau. There are, no doubt, many other modifications—as the Rarotongan au and kau both in Tongan and Hawaiian (Buschmann), tau in Nukuhivan (Mosblech); but these are sufficient to show the resemblances between the Dialects in this particular. The cases in Polynesian (if, indeed, there be any, which Dr. Maunsell, I think rather unnecessarily, calls in question) are clearly indicated by prefixing various particles, generally prepositions. Thus the Nominative is denoted in Maori, Tongan, Rarotongan by ko, and, in Tahitian, Hawaiian and Nukuhivan, by o. This particle is found before the article te, * The older form of the Island names Otaheite, Owhyhee, illustrate this; being O Tahiti, O Hawaii—so, also, the native name of the island Dominica—Hivaoa—which is written by Marchsand, Ohivaroa (O Hivaroa), and by Krusenstern, Ohivaoua (O Hivaoua). and the possessive pronouns, as well as before Plural particles, which precede the substantive. According to Buschmann, o occurs sometimes in Nukuhivan texts to mark the Accusative. The Genitive is, usually, shewn by the prefixes of no, na, o and a. Of these, o is the most common; while a is used, in a restricted sense, before living things. To and ta, also occur, and, in Hawaiian, ko, ka. The same pairs of prepositions serve to form the possessive pronouns by union with the personal. M. Buschmann remarks that those with a generally indicate a dependent, those with o an independent relation, and, further, that the genitival form in a, o, na, no follows the governing word, while ta and to precede it. In all the Polynesian languages, when two nouns come together without any particle between them, the second is considered to be in the genitive case. The Dative is very regular. Thus, in Maori, Tongan, Rarotongan, and Mangarevan it is shewn by the prefix of ki; and, in Hawaiian and

Tahitian by that of i only. Before Proper names of Persons, and before the Personal pronouns, ki becomes kia in Maori and Rarotongan, as, in Tongan, gui, becomes guia. M. Buschmann points out that, after the verb to give, the Dative of the person is expressed by the preposition na and no in Rarotongan, Tahitian, and Hawaiian, and that he has also detected na in Marquesan, as, a tuu maï na matuu, give us our bread. The Accusative, when marked at all, is, generally, shewn by a preceding i, in, at least, Maori, Tahitian, Hawaiian, and Rarotongan—and this i likewise becomes ia before proper names. In a great many cases, the substantive alone after the verb is sufficient, as, in the instance, from the Nukuhivan given by M. Buschmann, apevau te nata, call the man. The Vocative is marked in Maori, Rarotongan, Tahitian, Hawaiian, and Marquesan by é, preceding the noun. Occasionally in Tahitian and Hawaiian, é is found after, as well as before the substantive. The Ablative is indicated by i or e, the first being used to express means, cause or manner, while e is more usual after a passive verb. Taking next— III.The Adjective. The most general of all rules relating to the Adjective is that it follows the substantive and has not of it self, any distinctions of genderor number. It is usual to prefix ka when the adjective is alone, as ka roa, long; ka poto, short; but, when with, that is, after a substantive, the ka is omitted, as, he rakau roa, a long tree. Comparatives and Superlatives are formed by particles; by one or more words prefixed or post fixed; or by a repetition of the adjective itself, as, he waka nui, he waka nui atu, he mea nui nui. In some cases, the comparison is shewn by the insertion of the particle i, as, nui atu tenei waka i tera. This canoe is bigger than that—or without the i, as pai rawa te hanga a te tangata nei, this man's performance is best; in both of which latter instances, the adjective precedes the noun. The Plural is, being generally, made by a reduplication of the first syllable of the adjective—thus, he rakau pai, a good tree; he rakau papai, good trees: Archdeacon Williams however, remarks that these changes or modifications are not invariable, and that the simple form is often used in the plural. The reduplicated one, however, is restricted to the Plural. He adds, that the result of doubling both syllables of the roots is to diminish the intensity of the meaning of the root; as, wera, hot; werawera, warm (Dict., 1871, p. 8). In Tongan, there are some modifications. Thus, fu, great, very; foe, whole, entire, etc.; fu lahi, very many;—come before the substantive,—indeed, are used to a great extent, adverbially—hence, we find, fu ita, great anger (i.e. very angry);

Fu ulu, a single head—or the whole head; Fu ufi, a single yam, or an entire yam. Again, an interchange not unfrequently occurs—substantives being used as adjectives or vice versa, as, he vaka Fiji, a Fiji Canoe. In some cases, adjectives are derived from substantives by adding ia or ea—as, mafanna, heat; mafannaia, hot. In degrees of comparison, Tongan is nearly the same as Maori—but, the substantive verb would seem always to come first, as, gua lille ange he mea koeni gi he mea koia—is, better this than that thing (Mariner, p. 12). The superlative is, generally, made by the addition of the word obito most, very—as, this axe is the best, koe togi koeni gua lille obito. Dr. Maunsell observes that in Maori, adjectives, generally, take the form of the noun with which they are connected—i.e. if the noun be verbal, so is also the adjective—as, oranga tonutanga, Eternal life; that, when there are several qualities, the noun must be repeated with each quality; as, he tangata nui, he tangata pai, etc., that a common mode of denoting inferiority of degree is to associate together two contrary qualities as, pai kino, good—bad, i.e., indifferently good; Roa poto, long—short, i.e., of moderate length, etc., and, lastly, that, to represent the superlative degree, the definite article is sometimes prefixed with or without some word of intensity, as, ko au te kaumatua, I am the eldest son; ko te nui tenei o nga rakau katoa, this the large one (i.e., the largest of all the trees). M. Buschmann remarks that, while the determining adjective comes after the substantive, an adjective preceding, it acts as its attribute, in connexion with the verb “to be.” Thus, in Tahitian, te muua roa is, the high mountain, while é roa te muua means, the mountain is high (p. 173). Again, in Tahitian the comparative is made by the particles aé (ange in Tongan) and atu (so in Maori) placed after the Adjective. Mr. Logan has suggested (“Journ. Ind. Arch.,” Vol. V., p. 219), that the system of reduplication so prevalent in the Polynesian dialects is due to a love of “euphonic echo,” and that, by this means, plurality, intensity, repetition and reciprocity, are very effectively expressed. The same system prevails in Malay, and, occasionally, in Javanese, but it is far more extensively used in the Polynesian dialects than elsewhere. Thus, in Samoan, lau utele, is, a great tree; lau utetele, large trees. Rarotongan, Maki, sick; maki maki, sick persons. Javanese, homa gede, large house, homa gede gede, large houses. Again the Superlative in Viti (Fiji) is made of the reduplication of the Adverb, thus levu sara sara means, very very great. In Javanese, the same effect is produced by doubling the adjective—as duwor duworre, the highest. So, too, adverbs may be doubled, as genti genti, by turns. For intensity of action, we find in Tongan, tete, to tremble; tete tete,

to tremble much. So Rarotongan, kati, to bite; kati kati, to bite much. Hawaiian, lawe, to take; lawe lawe, to handle. Again, for repetition, plurality and reciprocity, we have in Maori, inu, to drink; iinu, to drink frequently. Tahitian, amaha, to split, amahamaha, to split repeatedly. Hawaiian, lele, to jump; lelelele, to forsake, repeatedly, (as a man his wife.) Nukuhivan, pepi, to strike; pepehi, pehipehi, to strike hard and often. Samoan, fete, to fear; na fefefe i latau, they were afraid; moe, to sleep; momoe, to sleep together. Tongan, nofo, to dwell; ke mau nonofo, to dwell together. Rarotongan, tae, to come; e tatae atu ra raua, and they two arrived. Tahitian, Taoto, to sleep; taotooto, to sleep together. In Javanese, hambedil bedil, means, continuing to shoot; bali, to return; bali bali, always returning. Vitian, ravu, to kill; sa vei raravui, they are killing one another. So, again, in Vitian, kamba, to climb; kamba kamba, a ladder. Tongan, lolo, oil; lolo lolo, oily. Rarotongan, paka, a stone; paka paka, stony. IV.Numerals.—In dealing with these, I am in great measure relieved from further labour, by Mr. Thomson's learned and admirable paper on the “Barata Numerals” and though, as I have stated, I cannot accept the special view which he advocates, I, at once, bear most willing testimony to the great ability shewn, not only in this paper (“Trans. N. Z. Inst.,” Vol. V., p. 131), and, in his two other papers “On the ‘Whence’ of the Maori,” which I have also noticed previously. Both these papers I have read three or four times over. I am, however, afraid, that, within the limits of my present paper, I shall not have space for any further examination of the questions Mr. Thomson has brought forward and discussed, but I may do so, hereafter, if I am able to procure the necessary addition to my at present, very limited supply of materials: meantime if I differ from him now, and may do so still more, if I ever have time, as I hope I may, to examine all his arguments as fully as I am sure they deserve to be examined, I rejoice to recognize in him a man, who has done, in the branch of Philology to which he has given his attention, first-rate work. I purpose, therefore, here, merely to point out what seem to me the chief characterictics of the Polynesian system of numeration, reserving for the present, any further discussion of Mr. Thomson's “Bharata” Theory. With reference to thè spelling of the Numerals, I have, in the case of the Maori, taken the forms given in Dr. Williams' most recent dictionary (1871); and, for the other dialects, that I have found most common in the books I have before me. The following table gives the leading forms:—for those of the Marquesas, Gambier and Hawaiian, I am indebted to M. Mosblech; for the Tahitian and Hawaiian to M. Buschmann compared with Mr. Andrews, and Adalbert von Chamisso; and for the Tongan, to Mr. Mariner—

I. II. III. IV. V. VI. VII. VIII. IX. X. Maori tahi rua toru wa rima ono witu waru iwa ngahuru Tongan taha ua tolu fa rima ono fitu valu hiva hongofulu or ulu Rarotongan tai rua toru a rima ono itu varu iva ngauru Samoan tasi lua tolu fa lima ono fitu valu iva safulu, ngafulu Mangarevan tai rua toru a rima ono itu varu iva ngauru Tahitian tahi rua or piti toru éha ha maha rima or pae ono or fea hitu varu or vau iva ahuru Hawaiian tahi lua tolu ha tauna lima ono hitu valu iwa umi I shall, therefore, only remark here that it is usual to place before the numeral, certain particles; of these, ko, ka, e, and a vary least in their several uses. Thus, in Maori, ko is invariably used before tahi and pu or topu for one pair;—thus, ko tahi pu, is the usual form for one pair. Ko tahi, ka rua, etc., answer the question, how many, as here, one, two, etc. When necessary, the individualization taki taki is made use of, as, ka waru taki taki nga kete, there were eight baskets once told. Again, for distribution, tá taki is prefixed to the numeral, as, kia tátaki rua pu nga utu i te tangata, let each man have four payments. So, taki whitu, means, by sevens. Toko is used when speaking of persons only, up to nine; hoko for multiples of ten. E occurs before all the numerals except tahi, but is not so definite as ka. The ordinals are generally expressed by tua or whaka, as, tua iwa, whaka tekau, or by a cardinal with a definite article, as, ko te wha tenei, etc. In the other dialects, the arrangements of particles, etc. is nearly the same. Thus Hawaiian often puts hoo before tahi—and ma is, generally, used to connect the different numerals as, in— Rarotongan—ngauru ma tai Each meaning 10 + 1 = 11. Maori—ngahuru ma tahi Samoan—sefula ma tasi Tongan—hongofulu ma taha In fact, as I have said in the case of Maori, so, in other dialects, we find toa too, toka, used for numbering persons—as, Hawaiian too piti. Tongan, toka hongofulu. Samoan, toa safulu, etc. The ordinals are arranged on the same principle, as in Maori. Thus we find ko te wa for fourth; so in Tahitian, we have, o ta ha (or' eha or maha); and in Samoan, o te fa. Lastly, as Maori has tua ngahuru for the tenth,

so Rarotongan, has tu rua. Hawaiian, tua lua and tua tolu; and Tongan, tuo ua or tuo tolu. * I ought to add that Forster (in his notes on Capt. Cook's Voyages), puts before all his Numerals ebo, as, ebo dahai, 1; Ebo houa, 2. But, as bo or po is the usual word for night, this evidently refers to the habits of the natives, who count by nights, as we by days. In Tahitian, rui is often used for night. V.The Pronouns, of which there are five Classes, Personal, Possessive, Demonstrative, Relative, and Interrogative, are somewhat complicated, but, chiefly so, from the number of words, many of which, in English, we should consider had only a quasi pronominal value. As a rule, they exhibit, throughout all the dialects I have been able to compare, a remarkable similarity, moreover are mostly found in all the three numbers of the Singular, Dual and Plural. It will be simplest to refer to each class separately, so I take first the Personal Pronouns: † I am indebted for this list partly to the Maori Grammars of Kendal, Williams, and Maunsell, and partly to the works of the Abbé Mosblech and M. Buschmann. The following scheme gives the First Persons of Personal Pronouns in the different Dialects: — Singular. Dual. Plural. Maori hau, koe, ia maua, taua, korua, raua matou, tatou, koutou, ratou Rarotongan au same Mangarevan au same Tahitian au, oe, oia maua, taua, 'orua, raua matou, tatou, 'outou, ratou Hawaiian au, oe, ia maua, taua, olua, laua matou, tatou, 'outou, latou Samoan au, ou, oe, ia maua, taua, olua, laua matou, tatou, outou, latou ma, ta, lua, la ‡ I do not feel sure that the above is quite correct. Any how Tongan differs a good deal from the others Tongan te or oa, ger-ia mau, tau mo, rau. The chief point to notice here is that the Dual and Plural are formed by the addition of the numbers, two and three, to the radical of the pronouns; but the roots differ from the singular and the numeral is somewhat contracted. Thus— We two (inclusive) is in Maori, Rarotongan, Tahitian, Marquesan, taua; Tongan, guita uua; Hawaiian, kaua. We two (exclusive) in Maori, Rarotongan, Tahitian, Hawaiian, is maua; Tongan, gui ma uua. You two—korua, Maori, Rarotongan; orua, Tahitian; olua, Hawaiian; guimoua, Tongan. In these we see the respective additions, of rua, lua, ua (two).

The two (i.e., one and the other)—rana, Maori, Rarotongan, Tahitian; laua, Hawaiian; guinauua, Tongan. In the Plural— We (inclusive)—Maori, tatou; Rarotongan, Tahitian, and Marquesan, matoou; Hawaiian, makoou; Tongan, guimautolu. We (exclusive)—Maori, matou; Rarotongan, Tahitian, Marquesan, matoou; Hawaiian, makoou; Tongan, guimautolu. You—Maori, koutou; Rarotongan, kotou; Tahitian, Marquesan, outou Hawaiian, oukou; Tongan, guimotolu. They (masculine and feminine)—Maori, ratou; Rarotongan, Tahitian, ratoou; Hawaiian, eakoou; Tongan, guinautolu. M. Buschmann points out certain peculiarities in the above Tongan words, as, for instance, the use of gui, the preposition of the Dative, and mo. The Tongan also uses the pure form for three, namely tolu (Maori, Rarotongan, and Tahitian, toru; Hawaiian, kolu). The Rarotongan, Tahitian, and Hawaiian drop the consonant of the numeral, and the Maori makes a further contraction of the vowels. In Tahitian, aou of the Pronoun of the First Person Singular is abridged into ou, after the prepositions a, o, na, no, ta, to, ia (Buschmann, p. 181), the elision being denoted by an apostrophe; and the pronoun ia, both in Tahitian and Marquesan, generally combines with the o of the Nominative and Accusative, making oia; and, also, after the above preposition, takes the form na, preceded by an apostrophe, as, to ia becomes to'na, of him, his; ia'na, to him, etc., for ia ia. * It is a pity that more care has not been taken in the printing different Polynesian dialects. M. Buschmann observes that sometimes the pronoun of the second person is contracted in such a manner that you cannot tell whether it is for I or you, my or yours. The American missionaries, on the other hand, he adds, distinguish in Hawaiian thus; writing a'ou, na'ou for of me, mine, and aou, naou, of thee, thine. But the most remarkable thing in the Polynesian personal pronoun is the existence of two distinct forms in the Dual and the Plural, distinguishing those persons who are really subordinate to the speaker from those who are not. Thus Maua, we two, means, I and my associate, not you. Taua, we, that is, I and you. So Matou, we (not you); tatou, we (with you). Mr. Logan points out (“Journ. Ind. Arch.,” Vol. V., p. 231) that this remarkable idiom is found, also, in the Malay and Philippine languages. The speaker is considered as the centre of being and action, and, in agreement with this, the present and future tenses are, as a rule, indicated in the Polynesian dialects by the definite article. Mr. Logan also thinks, that, to the idea of personality, which has a great influence in crude national minds and languages, is due, also, the double form of the possessive; viz.,

that which is mine attributively—or to possess merely, being indicated by the vowel o, while, that which is mine, objectively, or instrumentally, to act on or with, takes the stronger vowel a. M. Buschmann considers the peculiarity of the Mo, in the Tongan second person Dual and Plural, is also due to Western Malay influence; Mou, in Malay and Javanese, signifying you, yours, as Mo does in Tagala. I ought to add, that Dr. Maunsell considers some of the Possessive pronouns in Maori, are declensions of the Personal pronouns, as Naku mine. Nau thine. Nona his or hers, Noku Nou Nana and, so, also, Ia, as I a ia, from him, her; and Maua, or Maua, for him or her. Mr. Williams further thinks, that, with the exception of taku, tau, and taua, they have been formed from the genitive cases of the Dual and Plural numbers of the Personal pronouns. Thus he considers, to taua kainga the abode of us two—is properly, te kainga o taua, the article te and preposition o having coalesced. The Possessives in Maori are— Pers 1st Sing. Taku, toku. Plur. Aku, oku. Ta matou, to matou. A matou, o matou. 2nd Tau, tou, to. Au, ou, o. Ta korua, to korua. A korua, o korua. Ta koutou, to koutou. A koutou, o Koutou. 3rd Tana, tona. Ana, ona. Ta raua, to raua. A raua, o raua. Ta ratou, to ratou. A ratou, o ratou. Generally, but, not invariably, taku, tau, tana are used, when speaking of any thing done by or proceeding from a person, while toku, tou, to, and tona apply to something suffered by, or, in the possession of a person. Thus, taku kakahu, means, the garment I am making; toku kakahu, that belonging to me, or which I am wearing. The resemblances between Maori and the other dialects are, here, well marked. Thus, the Rarotongan is almost identical with the Maori, while the Mangarevan has takoe, tokoe, and the rest the same. Some, like the Tahitian and Samoan, have a contracted as well as a longer form, as Tahitian, tau, tou, nau, nou, au, oe, etc.; Samoan, leau, loau, lau, lou, aau, oou, au, ou, etc. Generally, M. Buschmann's view holds good, that the genitives of the Personal pronouns, formed with the three pair of prepositions, a, o; na, no; ta, to; (in Hawaiian, ka, ko) expresses fairly, the possessive pronouns

in the Polynesian dialects. The selection between those six forms by each separate dialect takes place on the same principles which regulate the formation of the genitive; the pronouns formed by the prepositions ta and to being placed before the substantive, and taking the case sign o; those formed by na, no, a, or o, following the substantive. M. Buschmann adds, that, in his Marquesan texts, he finds tou (for my) and to (they), the latter also occurring in Tahitian. The Demonstrative Pronouns are preserved most completely in Maori, as, Singular. Plural. tenei this. enei these. tena that. ena these. tera that (further off). era those. tana that. au those. (referring rather to objects.) Rarotongan is nearly the same— teia or eie teianei reienei, and modified forms, such as te reira, taua nei, aua nei, taua ra, aua ra. Dr. Maunsell points out that tenei (and its branches) are derived from te and nei, and are resoluble, as, ho mai te mea na Give me that thing. or ho mai tena mea and, that ia is also sometimes used demonstratively, as, tona wenua kai ha ia, this is the very land of food. The Samoan has— lenei this. nei these. lena that. na those. lela that. ia those. lea lae The Tongan is more defective, and has only, heni, this or these; hena, that or those, wanting, therefore, the adverb and pronoun of the third person, with the usual prefix of ko or a; but Mr. Mariner observes that there is very little distinction between koheni aheni and kohena ahena, here, as in Maori, and, elsewhere, the particles may be separated, as, hê tangata na, that man there. M. Buschmann observes that the Tahitian téi, téie, éié, correspond, in meaning, with the French ceci, or celui-ci, that is, the person nearest the speaker; while tena agrees with celui-là, the person or thing furthest off. He states, too, that, besides tera, there is also a form vera, used only in the

Plural. I or néi as is obvious in the Rarotongan, teiénéi. In Marquesan (Nukuhiva) he mentions, i téié née, to-day. tééi ce, celui-ci téié néi The local relations are formed by the connecting the substantive with one of the three adverbs of locality, néi, na or ra, as, taua, taata, na or ra, this man. Malay and Tagala, especially, preserve similar local relations. Relative pronouns, as distinct from others, do not appear to be used in the Polynesian dialects, generally they are, in fact, supplied by the sense of the passage, or parts of other pronouns are used for them. Thus, in Maori, the personal pronoun is used for the genitive, as ko te tamaiti tenei nona te ringaringa i wera i te ahi, this is the child whose hand was burnt in the fire (Williams, p. xxii.); or by ai after the verb, as, kei hea te pukapuka i tuhi-tuhi ai koe, where is the letter that you wrote? Ai is sometimes similarly used, as, nana ahau i ora ai, his (was the effort) by which I was saved. In the other dialects, similar devices are adopted. Thus, in Hawaian, te; Tahitian, tei and otei; Rarotongan tei and ko tei are met with. In the case of the Interrogative Pronouns, it seems to me that it is difficult, in many instances, to decide whether some of them are not more properly adverbs. Those, however, usually given in the Maori grammars are:— Wai Who ? Restricted to persons, as, Ko wai tera tangata who is that man? Kowai K' wai Aha. What? Restricted to everything meaning kind. Tehea Which? With reference to things. Hehea Besides these, Dr. Maunsell notes kohea and pehea, and adds that plurality is sometimes obtained by using ma (as we have seen before), as, Ko wai ma ena, who are they? The use of tehea is seen in the sentence, Ko tehea o nga waka i pakaru? which of the canoes was broken? Ko ehea tangata au e ki nei? which men did you speak of? In Tongan, Mr. Mariner gives— Kohai, Ahai? Who? With the same distinction as in Maori, between— (1) Men. (2) Inanimate objects or brutes. Koeha Which? Koehe heha What? Ahai Whose?

As, Kohai tangata ko hena. Koe tangata kohena ahái? But, Koe togi ko ena heha? What axe is that? In the same way, ki heha means, to which? i he ha " by what? fi ha (for fi he ha). How many? M. Buschmann (p. 184) points out that, in Tahitian, the forms are:— Aha (possibly, the Malay apa, and Javanese hapa) answering to the the French quoi? pourquoi? é aha (in Rarotongan, ěaa) téi héa (derived from héa, when, which is used in Hawaiian adverbially as well as pronominally) vai the sign of the Nom. with reference to persons. ovai éhia, in the sense of “combien,” “how many?” (possibly the Javanese pira), as, éhia fare? How many houses? At Nukuhiva—he finds ovai ovai aïoa—quel est le nom? éha tééi—qui est celui-ci? As, éha te méa néi? Who is it? What is it? éha ta oe—What are you doing? The Samoan shows the greatest simplicity of forms, and has rejected even more consonants than Hawaiian or Tahitian. Thus, I find— ai?—who? As, ko ai tou ingoa? What is thy name? i ai—to whom? a—what? fea—which? o le fea—which, of more than one? fia—how many? é fia fale—how many houses? So grammarians, as Dr. Maunsell, make further divisions of the Pronouns into Distributive and Indefinite; but these seem to depend more on their position in the sentence than on any thing else, moreover, are also used for other purposes. Thus, he expresses each, and every, by the Demonstrative or possesive pronoun, or by the noun twice or thrice repeated—as, I tenei ra, i tenei ra, each day; ia tangata, ia tangata, each man. In the same way, he states that some other, or any, are, generally, denoted by tetahi, etahi, etc.

In Tahitian, M. Buschmann points out the existence of nearly the same forms, as e.g., etahi or vetahi, e fanu, tu fanu: we find, also, atoa (Maori Katoa) as taua mau mea atoa, all these things. Tu fanu, which he renders “quelques uns,” he thinks corresponds with the Malay ano and the Javanese hanou. E occurs also, he adds, in the sense of “autre,” and is compared by him with the Maori and Rarotongan ke, and the Tongan guihe: he notes, further, that ke means strange in the Maori and Hawaiian and different in Tongan; he believes, therefore, that it may be compared with the Javanese zedje, which bears the three meanings of strange, different, another. In the Tongan, I notice kotoa with the sense of all, which is clearly the same word as katoa; and, yet another form in Hawaiian, with the same meaning, a pau or a pau loa. These, however, can hardly be modifications of the former. In considering the question of the Verbs M. Buschmann states that it is the weakest part of the Polynesian system of languages, and that, though there are abundant particles more or less connected with it, they fail to determine with accuracy even the principal times; while there is, also, no sufficient distinction between many of the particles employed to denote the separate moods. In this, he agrees, mainly, with Dr. Maunsell, who considers, truly enough, that there are comparatively few verbs in Maori, in the sense in which we speak of those parts of speech in Classical, or even European languages, as the same word may very often be a verb, a substantive, an adjective or an adverb. At the same time, as Dr. Maunsell further remarks, there are, no doubt, certain verbs, which may be considered as primitive, and certain others which are as clearly derivatives, comprehending under the latter head the reduplicated and compound ones. Since, therefore, às a rule, there is no variation of the ground form to denote, respectively, number, person, gender, mood, or tense; the simplest plan will be to consider separately each of the formations, whether by prefixes or postfixes, which are usually held to denote such changes, although it may be quite true as Dr. Maunsell urges, “that there are but few absolute forms for determining tenses.” As, in the comparison of the various dialects I have as yet been able to examine, the Maori is generally the best preserved, I propose to take the Maori verb first, and then to shew as far as I can, wherein the other dialects agree or disagree with it. Now, it is generally accepted (Williams', p. 24), that the Maori verb may be divided into Active, Neuter, and Causative, each of these divisions having its own passive. Thus, (1), the active and simplest form is clearly seen in the sentence, e kite ana ahau te tangata, I see the man; (2), The Neuter, (as expressing, generally, quality

or circumstance), è moe ana te tamaiti, the child is sleeping; (3), The Causative, made by prefixing waka (generally to Neuter, though sometimes to Active, verbs), as, pono, to be true; wakapono, to believe; mate, sick; wakamate, to make sick. The Passive verb expresses the action of some agent, as, e kitea ana té tangata e au, the man is seem by me; and is formed, by the addition to the active base or ground-form of one of the following particles:—la, ngia, a, kia, hia, ina, tia, kina, na, ngia, mia, ria and whira; the particle selected for this purpose, being, chiefly, determined by the termination of the verb, though many of these passival endings are quite arbitrary in their use. Frequentatives (as we saw before, in the case of the Adjectives), are expressed by reduplications, as, kokoti, to cut; kotikoti, to cut into many pieces. Tense, is shewn by the use of verbal particles—adverbs, prepositions, pronouns, and the articles he and te placed in connection with the verbs. These verbal particles (which have no meaning in themselves), are e, ana, ha, kua, i, kia, hei, me, kaua, aua, and kei. Thus the Present is formed by ka preceding the verb, or by e before and ana after it; as, kia rere te kaipuke ki Tauranga, the ship sails to Tauranga, etc. The Imperfect by e and ana, and some word or words to show that the action was incomplete when referred to; as, kai penei inanahi e haere ana, at this time yesterday I was going. The future Imperfect by ka or e before the verb, by ka before the verb and ai after it, or by ai alone after the verb; as, ka tere te waka aianei, the canoe will be adrift presently. Akuanei ano riro ai te kaipuke, to-day the ship will be gone. The present and future, when formed by ka, are generally to be distinguished by the sense. The Perfect has kua and i before the verb; as, kua ara mai te Karaiti i te mate, Christ has risen from death. The Pluperfect, with kua before the verb, must be distinguished by the construction; as, ka penei inanaki kua tae matou ki Puketona, at this time yesterday we had arrived at Puketona. The second Future, also indicated by kua, must, in like manner, be ascertained from the construction; as, e kore e po akuanei kua tae ahau ki Hokianga, before night I shall have arrived at Hokianga. The tenses of the subjunctive moods are indicated by help of conjunctions and adverbs; as, kua kite pea ahau i taua tangata otira e wareware ana ahau, perhaps I may have seen that man, but I forget. The Moods are shewn—the Indicative by e; as, e ngaki ana ia te whenua,

he is cultivating the land. The Subjunctive by me; as, me kaua te marangai, etc., if there had not been bad weather, etc. The Infinitive by kia and kei, and the Imperative by a great number of different modifications, the chief of which are the employment of kia, the absence of any particles whatever, or the prefixing of e to the future. Generally a passive form is used for the Imperative, as, karangatia e koe te tamaiti, let the child be called by you. The Imperative, is also further indicated by the use of maku, mau, mana, me, etc., in which, though the verb retains its active form, it is clearly used in a passive sense; as, me karanga e koe te tamaiti, the child must be called by you. Dr. Maunsell thinks that the verbal particles have some correspondence with the auxiliary verbs in English; at the same time they clearly do not admit of the same varieties of application, while they cannot claim the rank of the verb substantive. He thinks, too, with M. Buschmann, that no tenses can be accurately defined except the Present, Past, and Future. The distinctions he shews between simple and compound tenses are much to the point, and his analysis of the Imperative in Maori is valuable for the accurate study of the language, but is too detailed for the purpose I have here in hand. In connection, however, with the verb, I ought to add that there are a considerable number of what are called Verbal Nouns—their general object being to secure niceties and distinctions of meaning. Thus, wanaunga is, relative; wanautanga, a birth; kiteanga, the opportunity of seeing a thing; kitenga, the act of seeing; wakanga, the carrying on a back; wakinga, a breaking, etc. But, condensed, though my notice has necessarily been, I believe I have said enough to shew the general character of the verb in Maori; I proceed, therefore, now, to compare with it the verb in other dialects. The Tongan verb is characterised by its simplicity and regularity. It has but three tenses—past, present, and future—denoted by the signs, gooa, na, and me respectively, and three moods—the indicative, imperative, and potential. The first has no modal sign, the second neither modal nor temporal. The Subjunctive is marked by the modal sign ger. The order of constructions in the Indicative, is first, the sign of the tense, then the pronoun, and lastly the verb; except in the third person singular of each tense, where the pronoun is placed last. In the Dual and Plural, the pronouns ua and tolu follow the verb. Thus, the Present is denoted by gua, as, gua te alu, I go; the Past by na, prefixed to all persons, except the first, where it is changed to ne and joined to the personal pronoun u, as, ne u alu, I went. (The second persons, all through, are shewn by the form ger, as, gu ger alu, thou goest, etc.) The Future is indicated by te, except in the third person, where

it is changed to e and sometimes to teune, the pronoun being omitted; as e alu ia or teune alu, he will go. In the Imperative, in the second person, either the pronoun koi or the subject of the verb comes after it, but, in the first and second persons of the dual and plural, the pronouns tau and mo come before the verb and those, which distinguish the numbers, follow it, as, alu koi, go thou; tau alu, mo alu, etc., let us go (you and I), go ye (ye two). The Subjunctive is shewn by ger applied to any tense, as, ger te alu, ger ger alu, etc. In Tongan verbs, it would seem that the dual and plural are not carefully distinguished, the indefinite Plural (without ua and tolu) being constantly used. The Infinitive in Tongan is hardly distinguishable from an ordinary noun. Three words, my, atu, and angi are in constant use, with the general sense of “give” when verbs, and of “to” or “towards” when prepositions; hence, they are to be used, accordingly, as the first, second, or third person may follow, as, my ia giate au, give it to me; teu atu ia giate koi, I will give it to thee; angi ia giate ia, give it to him or her. Clearly, their chief object is to imply direction. Thus, of a is, to love; but “I love you” must be rendered not by gua te ofa koi, but by gua te of a atu giate koi. They also form parts of compound words, tálamg, talatu, talangi. It is a very general characteristic of all the Polynesian dialects that they love to use passive forms where we use the active for correct translation, and we have seen how fully supplied Maori is for the purpose of indicating the passive. Nearly similar forms (but much fewer in number) may be found in the other dialects. Thus, the Samoan has, a, ina, ia, fia, ngia, and tia; the Rarotongan, a, ia; the Mangarevan, ia; Tahitian, hia; and the Hawaiian, a, ia, hia, tia. M. Buschmann observes, that the use of the passive is equally characteristic of all the Malay languages, and he thinks he can detect the hia of the passive in the Marquesan kuhia. He adds his belief, that the character of the particle ua in Tahitian and Hawaiian (kaua in Rarotongan, kua in Maori, and gua in Tongan) marks a direct connexion between these tongues and the Malay and Kawi; holding that it is the same as the djoua of Malay, (houga of Javanese, and the djouga of Malay and Kawi, in all of which it is an adverb, with the heterogeneous meanings, of “also,” “only,” “thus,” “already,” “again,” etc. To this variety of meanings, he thinks, with William von Humboldt, is due the indiscriminating fashion in which it has come to be employed in the Polynesian idioms as the temporal particle of the verb, alike for the present, the imperfect, the perfect, and the pluperfect, the general definitions of which are anything but clear. It seems, indeed, to act as an auxiliary, in all the tenses, even in the future. Thus, in Marquesan, ua marks the present and perfect. Again, M. Buschmann further points out, the peculiar use of

auanei for the future in Tahitian and Hawaiian, and shews that te (the article before substantive nouns) and one of the local adverbs néi (here) or ra (there) seem, as it were, to embrace the Polynesian verb, so as to make it resemble a substantive accompanied by a demonstrative pronoun. The te before, and néia after the verb, makes the present in Tahitian, and te and ra, similarly placed, makes the imperfect. Occasionally, too, the adverbs are combined together, as, te first, then the verb, and then néi ra. The particles e and i are the most frequent auxiliaries to the Polynesian conjugation; they are found with all the tenses, excepting that i is not used with the future: they are found alone, or, joined with an adverb of time or place, are attached to the verb. In Tahitian, he gives the following uses—é—verb—ai—forming the future (in Maori, perfect and future; in Hawaiian, imperfect and future; in Rarotongan, é……éi). i……ai—-present and perfect (same in Maori and Hawaiian;in Rarotongan i……éi). é……néi—(in Hawaiian, present and future; Rarotongan, present). i……néi—present (in Rarotongan, imperfect). i……na—perfect. é……ra—imperfect and perfect (Hawaiian, é……la, present). i……ra—perfect. ua….ra—present and past. ra (alone after the verb), the past (so, la in Hawaiian). The particles e and i, exhibit a peculiar construction throughout all the Polynesian dialects, with the exception of the Tongan, viz., that the subject, when it precedes the verb, is combined with one of the three prepositions, marking the genitive, and terminating in a; as, a, na, or ta. When the subject is represented by a personal pronoun (see, before the forms, a, o; na, no; ta, to;) it takes the form of the possessive. But this construction is not found in negative or interrogative phrases. The imperative is indicated in Tahitian by é (as in Maori and Hawaiian), a (the ka of Maori and Rarotongan), ia (the kia of Maori), or éi placed before the verb; the last two particles being, in fact, the conjunction “that,” and the preposition indicative of motion. A joins itself with na after the verb. In Tahitian, as in Maori and Tongan, the verb, without any special sign, serves for the imperative. The Marquesan has the same simplicity of mood, with the use of é, as, noho oé, sit down (you, singular); noho, sit down (you, plural). The conjunction ia expresses the third person, as, ia tapu to oéinoa, hallowed be thy name (Maori, kia tapu tou ingoa). The prohibitory particles of the imperative, are, in Tahitian, auna (oua in

Tongan, kaua and aua in Maori) and éiaha; in Marquesan moï is used for the same purpose, and mai in Hawaiian. The Infinitive is denoted in Tahitian and Hawaiian by é before the verb; and the past participle of the Passive is shewn by i, preceding the passive form of the verb itself, as in iritihia, “translated,” (v. Tahitian Bible of the British and Foreign Bible Society). The English substantive verb, which is not generally rendered in the Polynesian dialects, is, in some degree, represented by the Tahitian temporal particle ua (the ua of Hawaiian and gua of Tongan), and by the pronoun of the third person, ia. M. Buschmann further shews that the Maori whaka (Tongan, faka; Rarotongan, aka; Tahitian, faa or haa; Marquesan, haka or haa; Hawaiian, hoo, sometimes haa) is, by no means, ong, the sign of the Causative verbs, but is found, not only with transitive and intransitive verbs, but, also, with substantives, adjectives, and adverbs. * Thus, in Tongan, faka (mode or manner), and ange (like or similar to), are joined to adverbs, etc., the former to verbs and adjectives, the latter more strictly to adjectives. The first is, as in the other dialects, always a prefix, the latter always a postfix. As, toa, brave; faka-toa, bravely; mamafa, heavy; mamafa-ange, heavily. There seems no sufficient principle for the employment of this prefix, and, possibly, all that can be said about it is, that it partakes the vague indeterminate character of a large number of other Polynesian particles. But, besides the particles connected with the verbs, to which I have already called attention, there are some others to be noticed, which M. Buschmann calls “particles de direction,” and which are variously employed. Thus, in Tahitian, two of these particles are directly opposed; mai (found in all the dialects in the same sense), “this way,” “towards me,” and atu (the same in Tongan and Maori, adu in Rarotongan, and aku in Hawaiian), “that way,” “towards you,” etc. There are two other particles aé (the same in Hawaiian, ake in Rarotongan, angi in Tongan) in the sense of “towards a third person,” and iho (the same in Hawaiian, io in Rarotongan), hifo in Tongan) in the sense of “downwards.” “Particles of direction” are employed after certain adverbs, whether simple or formed by a preposition, but their principal business is to accompany the verb, before the temporal adverbs, néi and ra, which are attached to it. Some other adverbs, however, which determine the character of the verb, as that of the passive hia, and the termination of the substantive raa, occupy the first place after the verb, and are, therefore, themselves, followed by the particle of direction. In the Marquesan, mai and atu are similarly used. Maori does not use ange. As mai essentially

belongs to the first person, we are prepared to find it constantly joined to the oblique cases, as, in Marquesan, ua tuu mai Jesu Mesia, Jesus gave it to us. In Nukuhivan, apéa maï oé, answer me; tuku maï, give me; mamui maï, follow me (Buschmann), in which latter case, it is equivalent to “here.” The simplest conception of atu, on the other hand, is that it belongs to the second person, as, in Marquesan, é nonoï atu au ia oé, I pray you. It is hardly necessary that I should prolong this paper by any detailed examination of the many other Particles in general use, whether as Adverbs or Prepositions, the more so that I could not presume to write a disquisition on Polynesian Grammar, and have no object in view but to point out sufficient similarities or diversities among the different dialects to enable me to draw some conclusions as to the supposed or real connexion between the existing inhabitants of these islands. For the same reason I abstain, altogether, from any discussion of questions of Syntax, which could not, indeed, be examined with any advantage without far more data than I at present possess. I may hope to do so some day. With regard to both adverbs and prepositions, I may, however, observe, that many of the most important have been incidentally noticed in earlier parts of the present essay. Generally, it may be said of the adverbs, that almost any word may become such, by the mere fact of being placed after the verb, but that a large number of them, as Dr. Maunsell has remarked, require some preposition to exhibit their application; many, also, are derived from words belonging to other parts of speech, while some are scarcely adverbs at all, in our sense of the word, but, rather, periphrases. Dr. Maunsell exhaustively groups them under the several heads of adverbs of time, place, order, quantity, quality, affirmation, negation, comparison, interrogation, and intensity, thus shewing that in these, as in other matters relating to grammar, Maori is much more rich than any of the other dialects. Perhaps, however, we are led to think so, in some degree, from the fact that the language of New Zealand has been more minutely and carefully studied than even Tahitian or Hawaiian. The latter, in its vocabulary, is considerably fuller. M. Buschmann points out that there are a considerable number of Polynesian words, which, by the use of prepositions, vibrate, as it were, between the substantive and the adverb; thus, preceded by prepositions, they express adverbs, but are, in fact, local and temporal prepositions; sometimes, also, they have another preposition also following them. Thus, in Tahitian, roto (same in Maori; Tongan, loto; Hawaiian, loko), as, i roko, within;i roto i, ěi roto ia before a personal pronoun; těi roto i, in;maï roto maï, out of; i roto pu i or ia, within. Again, ore—(not

kore, in Maori and Rarotongan;korang, in Malay; kaurang, in Javanese;kolang in Tagala)—is treated as a Verb in Tahitian and preceded by the particles é or a. The Adjective roa (same in Maori; loa, in Tongan and Hawaiian; dhava, in Javanese;lava, in Malagasi). “Long” is used in Tahitian for “very” and placed before the Adjective;so, also, ino, bad;and in Hawaiian and Marquesan, nui, great, as, oa nua, very high. On the subject of the Prepositions, I will only add that the following seem to be the simplest forms of them in Maori:—E, i, ki, hei, no, na, mo, ma, hei, o, a, ko, to, and that most of them may be recognised in the other dialects; as, in Tongan and Rarotongan—a, e, ki, i, o; mo, na, no, in Mangarevan, etc. M. Buschmann shows that in Tahitian i is employed in many and various ways—the same relations being found also in Marquesan. He considers it represents the i of the Hawaian, the ki of the Maori and Rarotongan, and the hing of Javanese; it is applied in Tahitian to all times, while é and éi have also a general similarity of sense;on the other hand, a is restricted to times future, and na and i na to times past, as is the case also with ne and ine in Hawaiian. I here bring to a close the few observations I have thought it worth while to make on certain forms occurring in different Polynesian dialects, and, while I am sure that they admit of almost unlimited expansion, I venture to hope that these will be considered sufficient to determine the question that the leading Oceanic dialects—the Maori, Tongan, Tahitian, and Hawaiian—are the remains of one original and wide-spread language. It now only remains that I should attempt to draw some conclusions from the evidence adduced in the previous pages, so far at least as this seems to point to the ultimate origin of the Polynesian population. Now, I think it will be admitted that, whenever I have found in such books as I have had the opportunity of examining, any apparent connexion between the Polynesian and other peoples, I have, in all cases, endeavoured to notice them. Thus I have repeatedly called attention to similarities existing between the Malay languages and one or more of the Polynesian dialects, with this principal object, that I might confirm, as far as I could, the evidence brought forward by Mr. Thomson, in the Appendix to “Trans. N. Z. Inst.,” Vol. VI., before referred to. That there is some connexion, I do not suppose any one can doubt who will take the trouble of fairly considering Mr. Thomson's arguments. The question is, how has this arisen? Have we, in short, any further reasons to support this connexion, and is what Mr. Thomson has urged, sufficient to enable us to say, unhesitatingly, that the Polynesians are Malays? I hardly think so;for what we are sure of amounts to

scarcely more than this—that in certain Malay, or so-called Malay languages, some grammatical forms, and also a certain number of individual words are found, both of which are also met with in Polynesian, though, in most instances, under considerable modifications of form. Clearly if we cannot say that languages so near akin the one to the other as Greek and Sanskrit can be placed under the category of derived languages, still less can we assert this in the case of Polynesian as compared with Malay. I am rather inclined to think that all that can at present be reasonably affirmed on the subject is, that there must have been a time when these two populations (the Malay and the Polynesian) were living near together, probably in intimate connection, and, further, that a long interval of time has elapsed since this occurred, during which there has been—almost certainly—an intervention of other races wholly diverse from both. Taking into consideration all the available facts, I think we are justified in believing that Malay and Polynesian, alike, ultimately came from some part of Central Asia, though, even here, I must admit that we have hardly anything that can be called evidence, and that it is only guess work as to the line or lines they may have taken from Mongolia to the Western Pacific. As an hypothesis I would suggest it is likely that, as we know was the case with the great waves of emigration, which at a period probably more recent, proceeded westwards across Asia into Europe, there were several routes eastward also, distinct the one from the other, but all, in the end, reaching the ocean. The originally one people, thus divided, might, perhaps, never again have met till long after they had occupied the island abodes where we now find them. Such a separation would be amply sufficient—on the analogy of what we know has happened in the progress of the Aryan (or Indo-European) races—for all the modifications of speech now noticeable in the dialects to which I have referred. There is nothing unreasonable in the supposition that what we can trace in the instance of the wanderings westward of the races of Central Asia should be equally true of other wanderings, in this instance, to the East and the Pacific, even though we cannot trace back these migrations with the same clearness that we can those to the West. The same reasons that led to migrations in the one would avail to produce the other;the most probable of these being over-population, and scant provision of food and of other necessaries of life. On this hypothesis it seems to me probable that there might have been two principal waves of emigration Eastwards, one finding its course along the great river highway of the mighty Yang tze Kiang, and thus reaching the ocean in the latitude of 32° N., with, possibly, a smaller branch by the southern stream of the Si Kiang, or river of Canton, reaching in 23° N.

latitude. The upper and main division would thence have found a nearly connected chain of islands, as the Ladrones, Carolina, and Radach Chain, etc., and might thus ultimately have attained even the extreme distance of the Sandwich Islands. In the same way those of the smaller branch, by the river of Canton, might have reached Luzon and the other Philippine Islands, and, possibly, by the same Caroline Islands have passed on to the more Southern as well as Easterly groups of Polynesia, such as the Fiji, Tonga, New Zealand, Society, and Paumatoa groups. Of course, this view partakes altogether of the nature of a guess;but, so far as we know at present, I do not think there is any thing in it unreasonable. The second main wave of emigration Eastwards, or rather South Eastwards, I suppose to have passed from Central Asia by the lines of the great rivers Brahmapootra, Irawaddy and Menam, thus impinging on the ocean at the South-east end of the Bay of Bengal and Gulf of Siam. These emigrants would, thence, naturally spread themselves in the direction of Tenasserim, the Malay Peninsula, Sumatra, Java, Borneo, etc., thus forming the ancestors of the present Malay races, though, as it seems most probable to me, at a period long antecedent to the 2,000 years, to which we are able, historically, to trace up the existing Malays. Now, if this theory of two or more lines of emigration has any consistency, and it be true that both Malays and Polynesians did come—it matters not how many years ago—from one original Asiatic source, some certain forms of speech, once correct in their ancestral homes, would be preserved by each wave as portions of a common heritage. The occurrence of similar grammatical forms, though perhaps few in number, would prove contact, if not relationship, at some period or other, while the absence of a large vocabulary of similar words would prove, also, a long and entire separation. We see the reverse of this in cases where the vocabulary is rich, but the grammatical words few or none. Thus, modern Turkish and Persian are loaded with Arabic words, but the one has not altered its Tatar or the other its Indo-European Grammar;on the other hand, France, which we know was once wholly Celtic, at the present moment, though still largely Celtic in race, has, with the exception of a few names of places, retained not one Celtic word in its spoken language. I am further induced to think that this view is confirmed by even the little we know of the Fiji and Tonga dialects, for which their grammar is sufficiently cognate with those of the other islands for their people to be generally included under the generic head of Polynesia. There are a great many words not Polynesian, and other words Polynesian originally but now altered (like tambu for tapu) to suit their organs of

pronunciation. * There is another hypothesis which, I think, ought not to be wholly discarded, and this is, that there has, at some period, been an emigration from America, westwards. If, as has been suggested, the idols on Easter Island have a considerable resemblance to those found in Mexico, it is not at all impossible that some of the earlier peoplers of Easter Island, or their kinsmen, may have reached Tahiti or even Tongatabu. Mr. Colenso, too, I see, thinks that the carving of the New Zealanders may be, perhaps, derived originally from America. Tongan has been clearly shown by Mr. Thomson to have many remarkable resemblances to Malay, and may some day prove to be an intermediate link by way of the Marianne and Caroline Islands, at least this I take to be the drift of M. Freycinet's researches. We have no historical, or even traditional, records on this subject; but a glance at the map suggests the probability that Melanesians from New Caledonia (the nearest Negro islands) may have found their way to Fiji, if not to Tonga. As both these populations were equally illiterate, the success of one over the other, if not the result of trade between them, must have been due simply to brute force: it would not have been like that of the letter-less Franks over the comparatively civilized and refined Gallo-Romans. I may add that the existence, both in Viti and Tongan, of many consonantal sounds, unpronounceable by any pure Polynesian, but at the same time not averse to the genius of other languages, point, necessarily, to such an intermixture as I have suggested; but when or how this came about, I doubt if we shall ever be able to determine. To recur to the native traditions: I have already stated the prevalent beliefs in New Zealand that the ancestors of the existing Maoris came from Hawaiki, and in all, or almost all, the islands a somewhat similar tradition is prevalent; in the Marquesas, indeed, the same name occurs unaltered. In general, however, this word has been slightly altered according to the consonantal system of each island, the varieties, according to Mr. Logan, being as follows:— In Samoa Savaii. " Tahiti Havaii. " Sandwich Hawaii. " Rarotonga Avaiki. " Nukuhiva Havaiki. " New Zealand Hawaiki. Captain Cook (Vol. III. p. 69) evidently refers to the same place in the name he writes, Heawige. Generally, it may be stated, that the popular idea is that this Hawaiki was somewhere under the islands—a sort of Inferno, confirmed

by the Tongan myth (prevalent also in New Zealand) I have already noticed—that their Chief God fished them up from the bottom of the sea. * I think I have seen it mentioned that, on some of the islands the tradition has died out or been forgotten; but that the word havaiki or avaiki has been retained with the simple meaning of “below,” “underneath.” It appears, further, that most of the islands place the residence of their Chief God in an island in the far West, called variously, Balotu, Salotou, and Purota. There is no island now to be found in that direction with any similar name, unless it be that of Bouro, a little to the East of Ceram. I have no means of telling whether Bouro contains any vestiges of Polynesian occupation; but, from its position, one would fancy it more likely that it would prove to be chiefly occupied by Negritoes. On the other hand, if it should turn out to be Polynesian, on the hypothesis of a descent from Central Asia, it would be well placed as a stepping-stone for further advance into the Pacific. The general inference from the universal occurrence of this word so little changed in form, and with nearly the same meaning, affords a strong argument in favour of the unity of the Polynesian race, though I am not sure that we can accept Mr. Logan's arguments for the order in which the different islands were peopled, because we find the name of Hawaii in both Society and Sandwich Islands, as well as in the most Western Samoan group. It seems almost hopeless, with such data as we have, to attempt any conclusion as to which island first used the word or the name; but if, as I think is certain, the migration was from West to East, it is reasonable to believe that the Navigator Islands might have had it centuries before it reached Tahiti or the Sandwich Islands, provided the migration took the course which I have called the Southern-eastern line. With regard to the great distances of water that the migrating canoes would, in any case, have had to pass over, there is certainly not the difficulty at first apparent, for Williams, in his “Missionary Enterprize,” having also clearly shown that there is not much more difficulty on the score of adverse winds. No doubt, over a considerable belt of the Pacific, East winds may be considered as the most prevailing; but not so as altogether to exclude the North and North-west, which often blow for days together. Kotzebue, in his voyage (Vol. II., p. 122) met with a native who had been driven 1,500 miles, with three companions from Ulea in the Caroline Islands, and who, as he had started from the East, still maintained that he had continued in that course; and, quite recently, a canoe was found 1,800 miles from its home; but the people in it were not starving, having caught fish enough for their support; moreover we know that, to this day, the Illanau people make annual voyages of more than 2,000 miles in quest of slaves and other plunder. I have before noticed that Tasman speaks of large double canoes as existing in his day on the coasts of New Zealand. Vessels of this

capacity seem to have gone out of fashion, at all events they are rarely now seen at any of the islands. Moerenhout, however, in his interesting voyage, states that he found such canoes in use among the people of the Paumatoa group, and with this unusual facility of construction, that they could be sailed whichever way their owners pleased by shifting their sails and rudder. Such vessels would, doubtless, have been quite fit to traverse very great distances. I believe that boats similar to these may be occasionally seen in the Fiji and Caroline Islands. I have already noticed that as you proceed from West to East it is a peculiarity of the Polynesian languages that they have fewer consonants, till at the Marquesas those are reduced to six, and it has been very generally asserted that this loss is a striking sign of degeneracy. But I am not so sure that this is a true view to take. Many of these changes, or rather modifications, are, I suspect due to climate, and certainly this is the case in well-known European examples. We may have a great, a natural respect for Highlanders; they may be, as they often are in our minds, the symbols of all that is manly, or brave, or virtuous; but it does not follow, indeed is not true, that the Italians, for instance, are as a body an effeminate race, though their language, from its vocalic character, lends itself more readily to love and music than the harsher languages of the North. Nor, indeed, is this true among the Islanders themselves. If the so-called effeminate Marquesans have only six consonants, the Maoris have but two more, and assuredly effeminacy could not be predicated of them as a race. Let us look a little nearer home, and see what has been done in the changes of the Old Classical Latin in the Romance dialects, and when we find in modern French such words as Augustus expressed by Aôut (only two vowel sounds, maturus by “mûr,” ligare, by lier, age (through étage) from œtas, let us not accuse even the Hawaiians or Marquesans as though a prevalence of vowels and a corresponding paucity of consonants was any proof of weakness in a language. Nor do I believe, as I have hinted previously, that, as a matter of fact, the Polynesian dialects are as deficient in vocal or consonantal sounds as we should infer that they are from the grammars and dictionaries already published. I suspect we have done the native languages much injustice, partly from the ignorance (not an ignorance worthy of blame), on the part of those who first reduced them to writing, of any principles of philology, and partly, also, from these varying sounds having been committed to paper by persons whose ears had not been accurately trained to the recognition of the niceties between sounds apparently similar. Had the Missionary Alphabet, drawn up chiefly by Professor Lepsius and Max Müller, been

available forty, or, better still, fifty years ago, I believe that even Marquesan and Hawaiian would have exhibited a list of distinct sounds, represented by letters; in other words, an alphabet which would have been little inferior to that of modern Italian. Anyhow, as I have already stated, the letter t would not have been banished absolutely from Hawaiian, and k substituted in its place, because a certain number of words occur in which the distinction between these two letters is not very rigidly preserved. I cannot help, also, thinking that, to express with perfect truth the shades of sound recognisable by a musical ear in Polynesian, it would be necessary to add letters from another language besides Latin, as, for instance, the θ for the English th.