ophryon, from the glabella, and from the nasion. In calculating the cephalic and vertical indices, I have, however, used only the second of the three, the glabello-occipital length, as recommended by M. Broca, Sir William Turner, and others.

Maximum Transverse Diameter.—This was taken as directed by Broca.

Glabello-iniac Length, Basi-bregmatic Height, Stephanic Diameter, Biauricular Diameter, Temporal Diameter, Minimum Frontal Diameter, Asterionic Diameter, Mastoid Height, Length and Breadth of Foramen Magnum.—These were all taken in accordance with Broca's directions.

Horizontal Circumference, with its Pre-auricular and Post-auricular Subdivisions.—In this also I have followed Broca, carrying the tape round the skull at the level of the ophryon in front, and the most distant occipital point behind. When I began this series of measurements I took a second horizontal circumference, from the glabella round the most distant part of the occipital bone; but, after an experience of about sixty skulls, I gave this up, owing to the irregularities introduced by the varying amount of projection of the external angular process of the frontal bone.

Nasio-opisthic Arc, with its Subdivisions—Vertical Transverse Arc, superior and inferior.—These were measured in the ordinary way between the points given by Broca. I found it impossible, however, to measure the inferior arc of the transverse vertical circumference at all accurately with a stiff broad tape, such as is used for the other curves, owing to the very irregular character of the undersurface of the skull. Broca recommends twisting the tape somewhat, so that it may pass between the styloid process and the spine of the sphenoid; but even with this precaution my results were very variable, and much in excess of the actual length of the curve. I found it easy, however, to take a satisfactory measurement with a piece of fine cord, and this was used accordingly.

Cranial Capacity.—In estimating the cranial capacity I have followed as closely as I could the directions given by Sir William Turner in his report on the crania collected by the “Challenger” expedition.^* I was, however, unable to procure the two-litre graduated glass used by him, and in estimating the quantity of shot contained in each skull I used the litre and half-litre measures used by Broca. Each skull was measured at least twice.

Basi-alveolar Length.—Here I follow Professor Sir William Flower.^† Unfortunately the central incisor region was fre-

quently broken, or altered by absorption consequent on loss of teeth, and in many skulls a true alveolar point did not exist. But, while even moderate absorption of the alveolar arch in front materially shortens those vertical measurements of the face which have their lower end at the alveolar point, the distance between the basion and the centre of the alveolar arch is not altered to the same extent. I have therefore sometimes given the basi-alveolar length in skulls whose ophryo-alveolar, naso-alveolar, and spino-alveolar lengths I have omitted.

Facial Angle (Ophryo-spino-auricular Angle).—This is the angle made by the ophryo-spinal and auriculo-spinal lines. It was measured by means of the goniometer designed for the purpose by Professor Broca, and made by Mathieu, of Paris.

Projection of the Zygomatic Arches.—This is shown by the terms “phænozygous” and “cryptozygous,” introduced by the late Mr. Busk. In estimating this feature the skull was placed on a Topinard's craniophore, with the condyles and alveolar point in one horizontal plane. If the zygomatic arches were visible to the observer's right eye placed at a distance of one metre vertically above the bregma, the skull was noted as Phænozygous. If they were invisible it was noted as Cryptozygous. In the tables the letters P and C are used to signify these conditions. When P alone is used, I mean that the arches, though visible, are not seen as free from the side of the skull. P + means that the interval between the arch and the skull can be seen from above.

Other Face-measurements.—These require but little explanation. They were all, excepting those of the palate, taken in strict accordance with Broca's directions. The palate was measured in the way proposed by Professor Flower,^* and adopted by Sir William Turner. The length given is the distance between the alveolar point and a line drawn between the most posterior parts of the maxillary tuberosities. The breadth includes the alveolar arch, and is measured at the level of the second molar tooth.

Indices.—The following are the indices selected, with the formulæ by which they are calculated:—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

In grouping the skulls according to their indices I have used the divisions as named and defined by Flower^* and Turner^†.

The grouping of the skulls according to their cranial capacities is as follows:—

I give in Tables I. and II. the measurements of eighty-three Maori skulls. Fifty are adult males, twenty-six are adult females, and seven are the skulls of children. Thirty-five of these are in the Canterbury Museum at Christchurch; twenty-nine are in the anatomical museum of the University of Otago; twelve are in the Colonial Museum at Wellington; while the remaining seven are in private collections.

The number of skulls of the Maori race already described by competent observers is sufficient to adequately establish the cranial characters of the race generally. I therefore did not undertake this series of measurements with

the object of possibly slightly modifying our knowledge on some points, but rather with the idea of ascertaining the craniometric expression of the differences which exist between certain tribes, and whether there is within the tribes themselves much individual variation in any of the more important cranial features. We know the Maoris to be a mixed race, the result of the mingling of a Polynesian and a Melanesian strain. The crania already examined leave no room for doubt on this point. But such skulls were for the most part collected either singly or in small groups from a very wide area—the whole of New Zealand, practically. Few, therefore, come from any one tribe, and, though their evidence as to the race-type, and its variations in the people as a whole, is ample, they tell us but little as to the characters of the smaller groups or tribes. So far as I have been able to ascertain, there has been but one contribution of any importance to the tribal craniometry of the Maoris. Sir William Flower gives in the Catalogue of the Museum of the College of Surgeons the measurements of twenty skulls, nineteen of which are adult, from near Whangarei. As these were found together in one cave it is probable that they all belonged to one tribe. They show considerable variety in form, but are, on the whole, rather long and narrow. I shall frequently refer to them in what follows. A much smaller group of six skulls found in a cave on the Island of Kapiti, and now in the anatomical museum of the University of Edinburgh, is described by Sir William Turner in his “Challenger” report; and to these may be added the two from the same island brought Home by the “Astrolabe.” Unfortunately two of the crania in Professor Turner's small series are not full grown; and the history of the island must make us careful of accepting as of one tribe, skulls collected at different times by different people.

With these objects in view, I have measured eighty-three skulls, forty-five of which belong to a single tribe. These last were all found in the South Island, in the Provinces of Otago and Canterbury. Though this is a very large area, it really supplies as satisfactory material for the study of variation within one tribe as can be got in New Zealand. A district of corresponding size in the North Island would yield individuals from several tribes, but in the South Island this is not so. In this Island, if we except the northern end, there is practically only one tribe, and skulls coming from widely-separated districts may therefore be made use of. To make this clear, I shall, following Mr. Stack's account,^* show how the South Island was peopled. We have first a tradition of a

tribe called Waitaha crossing Cook Strait from the north. Little is known of these Waitaha, but they are supposed to have come originally from the Bay of Plenty, and to have made their way south through the centre of the North Island. They are said to have spread themselves over the whole of the South Island, peopling it densely. This migration is held to have occurred in the latter half of the fifteenth century, but all such dates are of course most uncertain. The Waitaha held undisturbed possession of the land for at least a hundred years, when, about 1577, another band of invaders crossed Cook Strait, and soon conquered and destroyed or enslaved the peaceful Waitaha. These invaders, the Ngatimamoe, were an offshoot of the Ngatikahungunu, a powerful East Coast tribe, whose descendants still occupy the whole of the eastern half of the North Island to the south of Poverty Bay. After another more or less peaceful hundred years, the Ngatimamoe were, in their turn, called upon to defend themselves from the Ngaitahu, a second offshoot of this same Ngatikahungunu Tribe. The history of the Waitaha conquest was repeated, and the Ngatimamoe were absorbed by their more powerful relations. Still another invasion took place in 1827, when the Ngatitoa, a tribe from the west coast of the North Island, under Te Rauparaha, harried the settlements of the South Island Natives. But this was a mere raid, and did not result in settlement. The Ngaitahu still remain as the dominant tribe, and have given their name to the descendants of the conquered Ngatimamoe. The present Maoris in Otago and Canterbury are, then, though called Ngaitahu, the result of the fusion of that tribe with the Ngatimamoe. But these tribes, though hostile, were not distinct. They were really only sub-tribes given off at different times from the parent Ngatikahungunu stock. I describe forty-five skulls which belong to one or other of these two subdivisions. Their measurements are given in Table I. In Table II. are given the measurements of thirty-eight skulls from different parts of the North Island, a series obviously too small to be of any use alone towards answering either of the questions mentioned above. I include them in the present paper, however, from their bearing on the general type of the race, and also because they can be made use of to some extent when taken with those described by others. I hope that some day I may be in a position to make this North Island table more complete. In the meantime I divide the skulls into three groups, according to the districts where they were found. The first includes skulls from the East Coast, between Poverty Bay and the Wairarapa, the country of the Ngatikahungunu. In this group there are fifteen skulls, fourteen of which are adult.

Those in the second group were found on the shores of Cook Strait and the line of coast running northwards towards Cape Egmont. This contains ten adult skulls, one of which was found on the southern shore of the Strait, in the Nelson Province. I have, however, included it in this table because this part of the South Island was mainly peopled by Natives from the opposite coast. There are thirteen skulls in the third group, twelve adults and one child, from the Bay of Islands and the neighbourhood of Auckland.

In analysing these two tables, I give the results, as far as possible, in tabular form. In the first line of each of these smaller tables I take the skulls from Otago and Canterbury—Ngaitahu—then come the three groups of North Island skulls, while the last line gives the combined result of the entire series. Each line gives the average of the indices or measurements—(1) of the male skulls, (2) of the female skulls, (3) of those of doubtful sex, and (4) the general average of both sexes, with the extremes between which the individual skulls have varied. In each case the number of skulls on whose measurements the average is based is given. None but adult skulls are included in these tables of averages.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The average male index of the Ngaitahu skull is therefore mesocephalic, but almost at the upper limit of the group. The average of both sexes is of course much lower, but it is still mesocephalic. The range of variation for male skulls is 517c.c. Ten of the twenty-three male skulls, or 43 per cent., are megacephalic; eleven, or 47·8 per cent., are mesocephalic; two, or 8·7 per cent., are microcephalic. Of the North Island skulls, those which give the highest average are from the south-west coast, all the males being megacephalic. Those from the east coast have an average and a range of variation closely resembling the Ngaitahu skulls. Taking the combined results for both Islands, we find that the average of the forty-five male skulls whose internal capacity could be measured places them in the megacephalic division. Twenty-

six or 57·8 per cent., are also megacephalic; fifteen, or 33·3 per cent., are mesocephalic; while only four, or 8·9 per cent., are microcephalic.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

This table shows that the Ngaitahu come very low down in the mesaticephalic group, the male skulls being slightly longer than the female. The combined result for both Islands, however, shows this latter character reversed. The range of variation is 15·9, and between the two extremes the forty skulls from the South Island are arranged as follows: Three, or 7·5 per cent., are brachycephalic; twenty-two, or 54 per cent., are mesaticephalic; fifteen, or 37·5 per cent., are dolichocephalic. Even in much smaller groups the variation is considerable. For example, if we take the eight skulls collected in the small area in the neighbourhood of the Otago Heads, we find the cephalic index varying between 83 and 72·5. The average of the three North Island groups is dolichocephalic, though one of them is just within the mesaticephalic division. In none of these three groups is there a single brachycephalic skull. The most dolichocephalic group is that from Auckland and the Bay of Islands. Professor Flower's series of skulls from Whangarei, already referred to, and the three described by Sir William Turner from Auckland, likewise show a low cephalic index. The average of Professor Flower's—omitting the child's skull—is 73·5, and had he used the glabello-occipital length it would have been lower. Professor Turner's give an average of 72·7. We have therefore the measurements of thirty-four skulls from the northern end of New Zealand, not one of which is brachycephalic, and whose average index (even taking Professor Flower's at 73·5) is 73·4.

The average for the whole of New Zealand, according to my tables, shows the typical Maori skull to be at the lower limit of the mesaticephalic group: 4 per cent. are brachycephalic, 52·6 per cent. are mesaticephalic, 43·4 per cent. are dolichocephalic. Three have indices under 70. The seventy-

two crania referred to by Sir William Turner in his paper show the same small number of brachycephalic skulls, and have an average index of 74, a little lower than that given in this paper.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The Ngaitahu skulls are therefore metricoephalic. The range of variation is 15·1. Of the thirty-eight in the table, ten, or 26·3 per cent., are akrocephalic; nineteen, or 50 per cent., are metriocephalic; nine, or 23·7 per cent., are tapeinocephalic. The most northern of the North Island groups has again the lowest index, almost tapeinocephalic. The average for both Islands is the same as that for the Ngaitahu, and the grouping of the skulls is as follows: Fourteen, or 19·2 per cent., are akrocephalic; forty-four, or 60·3 per cent., are metriocephalic; fifteen, or 20·5 per cent., are tapeinocephalic. The general average also shows that the male skulls, following the usual rule, are slightly higher than the female. The comparatively low indices, which we have just seen to be characteristic of the Auckland skulls measured by me, are also shown by those examined by Professors Flower and Turner from the same district. The Whangarei skulls in the College of Surgeons Museum have an average index of 73·9, while that of Professor Turner's three is 71. The general average of the entire set of thirty-three crania is therefore 73·2. Comparing the cephalic and vertical indices, we find that in the Ngaitahu the average cephalic exceeds the average vertical by 1·3, and in the Auckland group by 0·5. In the other two groups from the North Island the vertical exceeds the cephalic—one case by 0·5, and in the other by 0·2. The general average gives a slight superiority—0·8—to the cephalic index.

In sixteen Ngaitahu skulls (nine males and seven females) the Basi-bregmatic height exceeds the maximum transverse width, while in twenty-two (fifteen males and seven females) the width is greater than the height. The greatest excess of width over height is in No. 6, Table I., where the height is 125mm. and the width is 141mm., giving a height-breadth

index of 89·3, The North Island skulls, on the other hand, have a majority whose height exceeds their width. In twenty-one is this the case, while in twelve the opposite condition is present. In two both diameters are equal. In these skulls the lowest height-breadth index is 92·8, the height in this skull being 128mm., and the breadth 138mm. Though the skulls which come from the southern half of the North Island are those in which the height exceeds the breadth of the brain-case most frequently, the numbers observed are too few and the district too large to allow of any tribal character being based on their measurements. Fortunately we are, thanks to Professors Flower and Turner, in a position to speak more definitely with regard to the skulls from the northern part of the Island. Combining their results with mine, we have, out of a set of thirty-three adult skulls, nineteen in which the height exceeds the width, thirteen in which the width exceeds the height, and two in which both diameters are equal. Contrasting these skulls in this respect with those from the South Island, we find that, while in 42·1 per cent. of the Ngaitahu the height is in excess of the width, there are 57·6 per cent, of these high skulls from this northern district. The average height of the Ngaitahu skulls is 136mm.; the average maxi-mum width of the same skulls is 137·3mm. These diameters in the thirty-three of the Auckland group average 136·2mm. and 136·3mm.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The proportion between the maximum width of the brain-case and its minimum frontal width as shown by this index varies little in the four groups. The range of variation is 14·7, and 59 per cent. of the indices are between 65 and 70. Comparing the asterionic and stephanic widths, we find that among the male Ngaitahu skulls there are eleven in which the asterionic and twelve in which the stephanic is the greater diameter. In two they are equal. Among the female skulls the asterionic is the greater in four, the stephanic in eight, and in two they are equal. The numbers in the North Island

groups are: Males—asterionic greater, eight; stephanic greater, fifteen; both equal, one. Females—asterionic greater, six; stephanic greater, five.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The large range of variation, 27, shows that the foramen magnum has no very fixed proportions. The length is equalled by the width in three cases.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The average index in the Ngaitahu group is mesoseme, and the range of variation in it is 23·6. Of the thirty-four Ngaitahu skulls measured, nine, or 23·5 per cent., are megaseme; fourteen, or 44·1 per cent., are mesoseme; eleven, or 32·4 per cent., are microseme. The proportion of female skulls is greatest in the lower groups, and this low orbital index in the female is worthy of note, as not being in accordance with the generally-received view. Two of the North Island groups have indices almost the same as the Ngaitahu. The east coast skulls, however, have a lower index, which brings them into the microseme division, but there are not enough of these skulls in the series to lower the general average for both Islands below 86·1, in the mesoseme group. Of the total sixty-eight skulls, seventeen, or 25 per cent., are megaseme; twenty-nine, or 42·6 per cent., are mesoseme; twenty-two, or 32·4 per cent., are microseme.

The average interorbital distance is 20mm.

The average orbital index of Professor Flower's Whangarei series, 91·9, is much higher than that of any of my groups. Professor Turner's three, on the other hand, have a lower index—80·3. Combining their results with my Auckland group, we have an average of 89·3 for these northern skulls.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The average index of the Ngaitahu skulls of both sexes places them in the lowest possible position in the mesorhine group. The males are also mesorhine, but the females are leptorhine. The range of variation is 19·2. Sixteen of these South Island skulls, or 48·5 per cent., are leptorhine; thirteen, or 37·4 per cent., are mesorhine; while only four, or 12·1 per cent., are platyrhine. The only leptorhine group is that from the south-west coast of the North Island. The east coast skulls are mesorhine, and those from the northern districts are higher in the same division. The general average for both Islands places the Maori skull very low down in the mesorhine division. Sixty-six skulls were available for the required measurements, and, of these, twenty-nine, or 43·9 per cent., are leptorhine; thirty, or 45·5 per cent., are mesorhine; seven, or 10·6 per cent., are platyrhine.

The highest average is, as we have just said, found in the Auckland group, where 25 per cent. of the skulls measured are platyrhine; but a small group like this, containing only twelve skulls, does not allow of conclusions being drawn as to the character of the nasal opening in the tribes of the far north. By making use, however, as before, of Sir William Flower's collection from Whangarei, and of those from the same part of the Island described by Professor Turner, we get a set of thirty-four, and these give a result but little different from that given in the table. Sir William Flower's skulls have an average nasal index of 49·9; Professor Turner's of 50·4; and the general average derived from their results and mine is 50·2 We get therefore a nasal index of 50·2 from thirty-four skulls collected about Auckland, and in

the region to the north; while my series of thirty-three Ngaitahu skulls give., as we have seen, an index of 48. It must be remembered, however, that the nasal index has been found to vary much even in races which are generally regarded as fairly pure.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

According to the averages all the groups are orthognathous, with the exception of the most northern, which is very slightly mesognathous. The range of variation is 12·9. Of the thirty-three of the Ngaitahu Tribe measured, eighteen, or 54·5 per cent., are orthognathous; twelve, or 36·4 per cent., are mesognathous; three, or 9·1 per cent., are prognathous. And of the sixty-two skulls which form the entire series, thirty-eight, or 61·3 per cent., are orthognathous; twenty, or 32·3 per cent., are mesognathous; while only four, or 6·5 per cent., are prognathous. The comparatively high index of the northern group seems to be the usual arrangement, for Professor flowers set, omitting the child's skull, has an average index of 98·2., and Sir William Turner's of 100; giving with mine an average of 98·4 for thirty-four skulls from this district.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

As the gnathic index shows the projection of the whole face, so this angle shows the degree of projection of the parts above the alveolar arch. This table, therefore, does not

closely correspond with the previous one, which deals with the gnathic index. The range of variation is considerable—14 degrees.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The averages place all the divisions in Sir William Turner's brachyuranic group. The greatest range of variation, 30·9, occurs as usual in the Ngaitahu Tribe, and the distribution of its thirty-two skulls is as follows: Brachyuranic, twenty-four, or 75 per cent.; mesuranic, six, or 18·8 per cent.; dolichuranic, two, or 6·9 per cent. In the south-west-coast group all the skulls are brachyuranic, and the index is highest. In Professor Turner's six skulls from Kapiti, however, one is dolichuranic, and three are mesuranic. The general average for the whole of New Zealand is, of course, brachyuranic. Of the sixty skulls in the tables, forty-four, or 73·3 per cent., are brachyuranic; thirteen, or 21·7 per cent., are mesuranic; three only, or 5 per cent., are dolichuranic. The measurements given by Professor Flower of his skulls from Whangarei do not permit of this index being calculated.

The width of the face relative to that of the brain-case is shown by comparing the bizygomatic diameter with the maximum transverse, and stephanic diameter of the cranium. In looking over the tables we find that in the great majority of the skulls the maximum transverse diameter is greater than the bizygomatic width. Among the male Ngaitahu skulls fourteen have the transverse diameter greater than the bizygomatic, while the reverse is the case in five skulls. Among the females of the same tribe six show the former condition, one the latter, while in another the two diameters are equal. In the combined North Island groups the proportion is twenty-two of the former type to seven of the latter, and in two the diameters are equal. The terms “phænozygous” and “cryptozygous” roughly indicate the relation between the bizygomatic width and the stephanic, or, rather, the maximum frontal diameter. As shown in Tables I. and II., all the male skulls from both Islands, with two exceptions, are phænozygous; and one of these cryptozygous skulls cannot be regarded

as normal in this respect, as a complete and early obliteration of the sagittal suture has checked materially the growth in width of its brain-case. In fifteen of these male phænozygous skulls, though the zygomatic arches are visible from above, the space between them and the bones of the temporal fossa is covered and invisible. In thirty the arches are seen free and open. The projection is less in the female skulls, and three of them are cryptozygous.

Basi-nasal Length, or Cranio Facial Axis.—The average length of this important dimension and its variations are shown in the following table:—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

In the paper on the cranial characters of the Fiji Islanders, already referred to, Sir William Flower gives a table showing the average length of this line in several races, both savage and civilised. In the male Esquimaux it is 106·1mm.; in the male Fijian, 104mm. In none of the other examples given does the length equal that given above as the average for the male of the Maori race, while the average for the female. Maoris of my series exceeds that of the females of all the races included in the table.

If we compare the basi-nasal length with the distance between basion and nasion, as measured over the vertex in the middle line, we find little variety in the different groups. Expressing the lower line as a percentage of the upper curve, we get for the Ngaitahu, 25·3; for the group from the southwest coast of the North Island, 25·5; for the east coast group, 24·9; and for the skulls from Auckland and the Bay of Islands, 25·2.

Professor Cleland,^* and, following him, Sir William Turner,^† make for purposes of comparison a somewhat different division of this mesial vertical circumference. Instead of using the

Characters of Cranial Vault.—In the majority of skulls the vertex is roof-like, with a more or less sharp and prominent median ridge in the parietal region, and a flattening, or sometimes hollowing, of the parietal bones between this ridge and their eminences. Some, however, have a rounded vault, and intermediate forms are met with. The outline of the longest skulls as seen from above is a very regular oval, the parietal eminences in these not being distinct. In most skulls, however, the parietal eminences are sharp and prominent, and, as they lie towards the back of the brain-case, the norma verticalis is obovate. Some skulls are pentagonal, as seen from behind. The forehead is, as a rule, moderately high and rounded, and any such pronounced flattening of the frontal region as is so characteristic of Moriori skulls is rare. A post-bregmatic depression was noted in seven skulls. The obelion was markedly depressed in two. The relative width of the parietal and frontal regions has been already described. The greatest transverse width is most often, in 36·9 per cent., on the parietal bones between the eminences and the lower border. In 29·8 per cent. it lies at the level of the parietal eminences, and in an equal number of cases at the squamous suture. I have noted it as being on the squamosals in two skulls, and in one on the suture between the mastoid portion of the temporal and the parietal. One skull alone showed the characteristic form due to early hydrocephalus.

The glabella is, as a rule, in the male skulls large and prominent, and the superciliary ridges long and curved. The temporal fossæ are large, and the temporal ridges usually run above or across the parietal eminences.

Sutures.—The coronal suture is generally simple, except for a short distance above the stephanion, where it is in most skulls deeply serrated. The sagittal suture, though not so simple as the coronal, is not complex. Nos. 2, 3, and 4 of Broca's scale would in most cases represent its serration. The lambdoidal is, as in Europeans, the most complicated suture. In one case it was, however, exceedingly simple, showing scarcely any serration.

The obliteration of the sutures of the vault does not seem to follow any very definite order, though on the whole these skulls lend support to the view that in savage races the closure begins anteriorly. In twenty-four skulls in which the obliteration of the coronal and lambdoidal sutures is partial it is further advanced in the coronal than in the lambdoidal in fifteen; in six the process is equally advanced in both; and in three the lambdoidal is in advance. The sagittal is in advance of the coronal in this respect in the proportion of eleven to six. In the sixteen cases which show partial obliteration of the sagittal suture the closure has begun

anteriorly in three cases only, while posterior fusion—the anterior end being free—is seen in twelve. In one the ossification has apparently begun about the centre.

In two skulls the sagittal suture has disappeared prematurely, not a trace remaining. One (No. 22, Table I.) is that of an adult; the other (No. 43, Table I.) is a child's skull, five or six years of age. They are both, in consequence, much more elongated than others of the same ages. The cephalic index of this adult skull is 69·6.

Metopism.—One skull is fully metopic. It was found in the Nelson Province, and is described with the North Island skulls. It is No. 11 of the group from the south-west coast, in Table II. In four other skulls a short fissure is to be seen in the glabella, and in a fifth a small triangular ossification is to be seen extending upwards into the glabella from immediately above the nasal bones.

Inter-parietal Bone.—No example of this irregularity was met with among the Maori crania, but in one case a small fissure 13mm. in length passed on one side into the occipital bone from its lateral angle.

Pterion.—The pterion is in almost all the cases in which it could be observed of the H-shaped type. The length of the parieto-sphenoid articulation varies from 3mm. to 20mm., the average being 9·5mm. In one specimen it seemed as if a K-shaped pterion had been present, but this could not be determined with certainty, because of extensive obliteration of the sutures. In one only is there an articulation between the squamosal and the frontal, and this, as in the Moriori to be afterwards noted, takes place by means of an oblong bony process which projects forwards from the squamosal as if an epipteric bone had fused with it. In a separate Maori temporal bone in my possession this same condition is present.

Epipteric bones are of fairly frequent occurrence. They occur in five skulls on both sides, and in ten more on one.

Wormian bones occur in thirty-seven skulls. In addition to the region of the pterion, they are found in the lambdoidal suture, where they occur most frequently, in the occipitomastoid and parieto-squamous sutures, and at the asterion. In one skull there occurs in each orbit a small separate ossification in the angle between the frontal, the malar, and the great wing of the sphenoid, and it is this skull which has the separate ossification at the root of the nose in the glabella already alluded to.

Incomplete union between the squamosal and petro-mastoid elements of the temporal bone was seen in two skulls, occurring on both sides in each. In these a fissure, such as has been described in European skulls, is seen running obliquely from the angle between the squamous and mastoid above, down-

wards across the mastoid process, nearer its anterior than its posterior border. In neither case is the separation between the two elements complete.

Paramastoid Process.—This was noted in eight skulls, in four of which it is present on both sides. As a rule they are small and unimportant, but those of a skull from near Dunedin merit a short description. They are both low and blunt, but are each articular; one, the left and the longer, articulates with the outer surface of the corresponding lateral mass of the atlas, while the other articulates with the upper and back part of the right lateral mass. This skull is part of a complete skeleton, and the atlas shows the facets which correspond to these paramastoid articular surfaces. In another skull there is a non-articular paramastoid on one side, while on the other there is, not a projection, but a smooth flattened area on the undersurface of the jugular process, which in all probability was for articulation with the transverse process of the atlas.

In one female skull, also from Otago, the atlas is fused to the occipital bone by its anterior arch and lateral masses.

Third Condyle.—A slightly-raised third condyle, 8mm. broad, is present in one skull close to the anterior edge of the foramen magnum. In another the anterior border of the foramen is articular, but not raised. On the contrary, it is slightly concave, as if the tip of the odontoid had pressed on the occipital. In a third skull, although there is no articular surface, there are two bony nodules lying close together on the undersurface of the basi-occipital, a very short distance in front of the border of the foramen.

Parietal Foramen.—This, though present in the great majority of skulls, is often exceedingly minute. It is absent in 20 per cent. of the crania examined. It is present on both sides in 32·5 per cent., and on one side in 45 per cent. One skull has two foramina on one side, the right; and in four the foramen is in the middle line.

Posterior Condyloid Foramen.—This also varies in size, but not to the same extent as the parietal foramen. It is absent in nearly 8 per cent. of the skulls measured. It is present on both sides in 56 per cent., on one side in 36 per cent., occurring more often on the left side than on the right side.

External Auditory Meatus.—No obvious diminution in the size of this canal was observed. No exostoses are present, but in one adult skull the tympanic plate is perforated.

The superior curved line of the occipital bone is clearly double in 33 per cent. of the crania examined.

Equilibrium of Skull.—Nearly 40 per cent. of the skulls, when placed on a table, rest on the teeth, and either the

postero-lateral edges of the foramen magnum or the occipital bone near the foramen. Thirty-four per cent. rest on the mastoid processes, a posterior mastoid equilibrium being present more frequently than an anterior. In a little over 20 per cent. the condyles project beyond the mastoids, and the skull rests on them, while in a few the lateral points of contact are one mastoid and one condyle.

Infra-orbital Suture.—Thirty-one per cent. of the skulls have this suture present on both sides. In 15 per cent. it is present on one side only. In about 10 per cent. of the instances observed it is partly obliterated.

Lachrymal Bone.—A pronounced deviation from the ordinary type was observed in only one New Zealand skull. The inner wall of the orbit is, however, frequently greatly broken. In the skull referred to there is no separate lachrymal on either side, the nasal process of the superior maxilla being prolonged backwards to meet the os planum of the ethmoid. A small space is, however, present in the lower part of this suture, and this was in all probability filled during life by an ossicle. The os planum is occasionally narrowed anteriorly, the shortest ethmo-lachrymal suture I have noted being 5mm.

In another skull, which I believe to be Maori, though it was found at the Chatham Islands, an irregularity also occurs in this region. The measurements of this skull are given with those of the others from the Chathams, in Table III., but the lachrymal irregularity may be described here. It occurs in the right orbit. In it the lachrymal bone and the os planum come in contact by a mere point. Below they are separated from each other by a triangular process of the orbital plate of the superior maxilla, above by a triangular ossicle which articulates along its third side with the frontal bone.^*

Spheno-pterygoid Foramen.—Traces of this are present in nearly 50 per cent. of the skulls of my series. In one it is complete on both sides; in two more it is complete on one side, and in one of these the bridge of bone which connects the pterygoid plate and the spine of the sphenoid is 7 ½mm. deep. It is nearly complete in six skulls, one of these having the same condition on both sides; it is more than two-thirds formed in nine more; while in nineteen a spine is to be seen projecting backwards from the posterior edge of the external pterygoid plate. In those skulls which

have the foramen complete, or nearly so, the pterygoid plate is very much broader than usual.

Nasal Bones.—The nasal bones show considerable variety in size and form, even in skulls from the same district; but in male skulls the bridge of the nose generally seems sunken below the prominent glabella. In a few it is here very narrow, and in three skulls, one nasal bone alone reaches the frontal. The lower part of the nasal bones is usually convex; and though in a number of skulls they are short, and in a few flat, yet in many the bridge of the nose is high, prominent, and curved.

Anterior Nares.—In one skull only is this opening quadrangular. Though it varies considerably in its proportions, it is almost invariably rounded, and much narrower above than below. The lower margin is rounded in 37 per cent., sharp in 31 per cent., two-lipped in 20 per cent., and bevelled or sloping gradually to the facial surface of the superior maxilla in 12 per cent. of the skulls in which it could be examined.

Alveolar Arch.—In nearly half the skulls this is of the ordinary parabolic curve. Of the remainder, the great majority are hyperbolic; while in only from 9 to 10 per cent. is the hypsiloid form of curve present. In none is the curve elliptical. In one, a female from the west coast of Otago, there is an incurving of the premolar and anterior molar region, so that the arch in this part is concave outwards.

Lower Jaw.—This bone is generally strong and massive. The angle is occasionally sharp and easily localised, but is more frequently rounded, and this rounding is in many bones carried forwards along the undersurface of the body, curving upwards anteriorly towards the chin. Indeed, in some cases the posterior margin of the ramus, the angle, and the lower border of the body form one long continuous curve, whose most dependent part is below the first molar tooth.

The coronoid height is greater than the condyloid in the great majority of the mandibles examined. In five cases it is less, and in one these two measurements are equal.

The intergonial diameter exceeds the gonio-symphysial length in every case.

Teeth.—In very few of the skulls in this series are all the teeth present. The loss has in most cases occurred after death, or in advanced age, but in several adult skulls there are undoubted signs of ante-mortem loss of teeth. These have probably been lost as the result of injury, as in no case have I been able to observe the slightest sign of dental caries. But though I have not seen decayed teeth in these Maori skulls, I have noted in seven the cavities of alveolar abscesses. Six of these cavities are found in the upper jaw, one in the lower; and most of them have been at the roots of either the

incisor or the premolar teeth. Professor Rolleston^* has noted this condition in several of the skulls from British barrows examined by him—skulls belonging to a people whose habits in many respects, in all probability, closely resembled those of the Maoris. Most of his instances occurred in female skulls. All mine were met with in males.

The third molars are, as a rule, not ground, or only very slightly so; and the second molars are usually not much worn; but all the other teeth, except in one or two skulls, show in a marked degree the wearing-down of the teeth characteristic of races in the condition of the Maoris, frequently the whole crowns having been ground away. In several skulls I have observed a curious condition of the first molar teeth of the upper jaw. In its slighter degrees there is merely an oblique wearing of the crown, the outer side being most worn. In the most pronounced cases, however, the pressure from without has caused a gradual dislocation of the tooth inwards, so that it comes to lie transversely in the jaw, with the two outer fangs exposed, and these, with the remaining portion of the crown, are ground down so that the pulp-cavity is exposed in its entire length. The chewing-surface so produced is not flat, but rounded from within outwards. The incisor teeth are quite as much worn as the first molars and the premolars, and, instead of cutting-edges, have flattened crowns. Unfortunately they have been lost even more frequently than the molar teeth, and in comparatively few cases have I been able to determine satisfactorily the relation that the upper incisors have had to the lower; but in some skulls it is clear that the upper teeth do not, as with us, project in front of the lower, but lie directly above them, the flatly-ground surfaces of each being in contact. This arrangement has been described by Sir William Turner in Australian skulls.^†

The molar teeth are generally three on each side, or, rather, have been so during life, as, of course, many have been lost after death, and many more, though once present, have been lost during life from old age or other causes. In fifteen skulls, however, I have been unable to find any trace of a third molar on either side of the superior maxilla, and in another, this tooth, though present on one side, has apparently never been developed on the other.

Children's Skulls.—The measurements of nine young skulls are given in the tables. Two of these were found at the Chathams; the others come from different parts of New Zealand.

from Hawaiki, some twenty-seven or twenty-eight generations ago. On arrival they found the islands thickly inhabited by natives who differed considerably from them, being darker, and having very black hair. After much fighting, they made peace with the islanders, and, intermarrying with them, the two races became fused. There is another tradition of the arrival of a second body of immigrants at a later date. These are said to have come from New Zealand.^*

This traditional history tends to show that the Morioris, like the Maoris, are of a mixed Polynesian and Melanesian stock.

One of the earliest of the English settlers^† describes them as “of middle stature, with almond-shaped eyes and hooked noses,” bearing “a most remarkable resemblance to the Jewish race.” I am told, however, by one of the European residents that this Jewish nose is by no means universally present—that, indeed, broad and rather flat noses are more common. They differed considerably from the Maoris in appearance, as they did not practise tattooing. In character, too, they were unlike the New Zealand natives, being gentle and timid, and totally devoid of the energy, intelligence, and ferocity of their conquerors.

The measurements of about thirty Moriori skulls have, I believe, been already published. Four crania, one of them a child's, were collected by the “Challenger” expedition. These, with five others in the anatomical museum of the University of Edinburgh, have been described by Sir William Turner^‡ in his report on the human crania in that collection. Eight crania are in the museum of the Royal College of Surgeons in London, and have been measured by Sir William Flower.^§ Three are described by Dr. Zuckerkandl,^| from the collection made by the “Novara” expadition, and one of these is figured. Five are in the Paris Museum, and are described by MM. de Quatrefages and Hamy;^¶ one of these is also figured. Dr. Barnard Davis^* gives the measurements of three more, one of which, however, is a child's. Some measurements of one skull are given by Dr. F. J. Knox,^† in describing a

Moriori skeleton in the Colonial Museum, Wellington; and a few crania are, I believe, in the Godeffroy Museum at Hamburg.

I describe in this paper fifty skulls from the Chatham Islands. Thirty-four of these are adult or aged males; seven are adult females. In three other adult skulls the sexual characters are very ill-defined, and I class them as of doubtful sex. Six are the skulls of youths or children.

Twenty-eight of these skulls are in the anatomical museum of the University of Otago; seventeen are in the Canterbury Museum; five are in the Colonial Museum at Wellington.

Though all fifty were, I believe, found at the Chathams, I do not regard them all as Moriori skulls. From what has been said above it is clear that Chatham Island skulls are not necessarily those of the Moriori aborigines. They may be so; but they may also be Maori, or even European; and in my collection it is easy to recognise different types of skull, and, though none of them are European, I have come to the conclusion that four skulls are Maori rather than Moriori. These differ from the others in several respects, but especially in the form of the cranial vault, and resemble more closely some of the skulls found in New Zealand, near the western opening of Cook Strait—that part of the North Island where the Ngatiawa and other invaders of 1835 lived before their exodus to the Chathams.

The other forty-six skulls are, in my opinion, Moriori, but not all of one type. Amongst the adult skulls two types may be recognised, and the skulls may be divided into groups according to their resemblance to one or other—groups, however, which shade into each other through intermediate forms. The typical members of the first group are usually large and rather heavy skulls, with prominent parietal eminences and roof-like vertices. They are all more or less pentagonal as seen from behind, some very markedly so, and the low flattened retreating frontal region is a most striking feature. The excess of width over height is generally well marked; indeed, in the most typical members of the group the brain-case is distinctly flattened. The orbits are, as a rule, high, and the appearance of height is increased by the form of the superciliary ridges; while the nasal opening is narrow, with long prominent nasal bones, which are convex below. The air-sinuses in the frontal bone are mainly confined to the region above the root of the nose, so that, while there is, as a rule, a massive prominent glabella, the superciliary ridges are short, and do not pass far out over the orbits. The majority of the skulls from the Chatham Islands that I have examined are of this type. The skull numbered 27, in Table III., shows a different type, and several others resemble it more or less

closely. These form the second group. The main differences between these skulls and those I have just described are the higher and more rounded forehead, the less pentagonal norma occipitalis, the less projecting jugal arches, and the smaller orbital openings. By the “smaller orbital openings” I do not mean that the orbital index is lower in these skulls. This is certainly not the case in all. I mean rather that the orbital opening relatively to the rest of the face is smaller in the skulls which compose this group than in those of the first. The more overhanging superciliary ridges do, however, take away somewhat from the appearance of vertical height. Some of the skulls in this group are also large, one having a capacity of 1,650c.c. These crania are of a more Maori type than those in the former group; but I class them all as Moriori, because I have a clear series of gradations between them and the most typical members of the other group, and also because I have been told by one of the English settlers that in undoubted Moriori burial-places skulls of both types are found. That there should be found among Moriori skulls some which might be regarded as Maori is almost to be expected, as there can be little doubt that both Maori and Moriori are the result of a mingling of the same races—in different proportions perhaps, but still the same. There is, besides, as already noted, a tradition of the early introduction of a pure Maori strain direct from New Zealand; and, remembering the nearness of the Chatham Islands to New Zealand, it is difficult to believe that this occurred only once in the long period that elapsed between the settlement and the arrival of Europeans.

The measurements and indices of these Chatham Island skulls, taken in the way already explained, are given in Table III. I give in the table the entire series, but the four skulls which I regard as Maori are placed in a division by themselves at the end. They are numbered 47, 48, 49, 50.

The short tables included in the text contain, as before, averages calculated from the figures in the larger table. They deal with adult skulls alone.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The average of the male skulls shows them to be megacephalic; the female average is microcephalic; and both

sexes combined give an average which is in the upper half of the mesocephalic group. The small number of female skulls makes the combined average rather unsatisfactory. Had the sexes been in more equal proportions the average would have been lower, and would have more correctly expressed the general fact.

The range of variation is 465c.c. if we include both sexes, 365c.c. if we take the male skulls alone.

Of the male skulls, fifteen, or 51·7 per cent., are megacephalic; eleven, or 37·9 per cent., are mesocephalic; while only three, or 10·3 per cent., are microcephalic. One of the females and one of the skulls of undetermined sex are at the lower limit of the mesocephalic group; the others are microcephalic.

Professor Flower has taken this measurement in seven Moriori skulls, and these have an average capacity of 1,396c.c. The average of Sir William Turner's series of eight is still lower, 1,387c.c. but in both groups the percentage of female and doubtful skulls is higher than in mine. It is also to be noted, as explaining my higher results, that the collections measured by these two observers contain none of the skulls of exceptionally large size described by MM. de Quatrefages and Hamy.^* The mean cranial capacity of the three males in the Paris Museum is given by them as 1,600c.c., of the two females as 1,565c.c., while one of the male skulls has a capacity of 1,785c.c. Sir William Turner's largest has a capacity of only 1,492c.c., Professor Flower's of 1,560c.c., whereas I show in my table three skulls whose capacity ranges from 1,580c.c. to 1,650c.c. The three “Novara” skulls have a mean capacity of 1,428c.c.; Dr. Barnard Davis's two adults of 1,475cc.

Combining these measurements with mine—in all, sixty-three skulls—we get the following results: Average capacity of forty-seven male skulls, 1,460c.c.; average capacity of sixteen skulls of female or of doubtful sex, 1,326c.c. The male Moriori skull is therefore low down in the megacephalic group. The female is microcephalic.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The male skulls, the female skulls, and the skulls of undetermined sex have all, therefore, average indices in the mesaticephalic group, the females having slightly higher indices than the males. The general average is also low down in the same division.

The range of variation is 11·1.

One skull only is brachycephalic. Thirty-one, or 77·5 per cent., are mesaticephalic; eight, or 20 per cent., are dolichocephalic; but no skull has a cephalic index under 70.

Professor Flower's series of eight has a higher average index—78; but, as already pointed out, his method of calculating the index leads to a higher result than that which I have adopted. My results are more directly comparable with Sir William Turner's. The eight skulls measured by him have an average index of 75·2. The three of the “Novara” collection average 75·9. The Paris skulls average 76·3; and the adults measured by Dr. Barnard Davis, 74·9.

Combining these results—exclusive of Professor Flower's—with mine, we get from fifty-eight skulls an average cephalic index of 76·1

The average glabello-occipital length in my male skulls is 187mm.; in the female, 175·6mm. In the males the maximum transverse diameter of the brain-case averages 142mm.; in the females, 134·2.mm; while the average of both sexes is 140·1mm.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

In all the groups the average index is metriocephalic, close to the lower limit of the group. The index of the female skulls is very slightly higher than in the case of the males, but the small number of female skulls measured makes it impossible to say whether the Morioris are really exceptions to the general rule in this respect.

The range of variation is 13·3.

Only two skulls in my set are akrocephalic, and these are both males. Twenty-one, or 51·2 per cent., are metriocephalic; eighteen, or 43·9 per cent., are tapeinocephalic. Five of the females belong to the metriocephalic group; the others are tapeinocephalic.

As the mean length-breadth, or cephalic index, is 76·3, we see that the average Moriori skull is distinctly lower than

it is broad. Indeed, if we look at the measurements of the individual skulls, we find that in only two (Nos. 22 and 29) does the height exceed the maximum width, and then only by two millimetres in one case and by one in the other, while in two skulls more these diameters are equal. In one skull (No. 11) this excess of breadth over height amounts to 18mm., giving a height-breadth index of 88.

In this respect the Moriori skulls differ distinctly from the Maori. As will be seen by referring to the earlier part of this paper, the average vertical index in Maori skulls is 74·6—slightly lower than the cephalic index, which is 75·4. In them there is an excess of the cephalic index over the vertical of 0·8. In the case of the Morioris, however, the excess is 3·6. It is true that there are several Maori skulls given in the tables whose vertical indices are lower than the Moriori mean, but there is no Maori skull with an index as low as the lowest Moriori, nor no Moriori as high as the highest Maori. Also, in over 90 per cent. of the Moriori skulls the width is greater than the height, while in only from 40 to 50 per cent. is this the case with the Maori skulls that I have examined.

The highest cephalic index and the highest vertical index belong to the same skull, and there is a general correspondence between these indices throughout the series.

Professor Flower's series have again a higher index than mine—73·9—because of the measurements selected; and Professor Turner's have an average of 73. I have not calculated this index for the “Novara” skulls, as the height given is not the basi-bregmatic; but the mean index of those in the Paris Museum is 72·8.

The average basi-bregmatic height of the male skulls that I have examined is 135·5mm.; of the females, 128·1mm.; while the combined average is 133·7mm., showing an excess of breadth over height of 6·4mm.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The range of variation is here 16·4

In my series of Maoris this index averages 68·1. The anterior narrowing of the cranium is therefore more pronounced in the Moriori than in the Maori.

In the forty skulls in which it was possible to compare

them, the asterionic diameter exceeds the stephanic nineteen times; the reverse condition occurs eighteen times; and in three skulls the two diameters are equal.

Though in the majority of cases the stephanic diameter is the maximum frontal, yet in fifteen of the skulls examined the greatest frontal width lies a little below the stephanion.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The range of variation is here 23·5.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The male average is high up in the mesoseme division; the female is low in the megaseme; and the combined average is almost in the latter group.

The range of variation is 24·1.

Of the forty-two skulls measured, twenty, or 47·6 percent., are megaseme; eighteen, or 42·9 per cent., are mesoseme; and four, or 9·5 per cent. are microseme. A much larger proportion (32·4 per cent.) of the Maori skulls examined were noted as microseme, while the percentage of megaseme Maori skulls was only 25.

The average female index is seen to be higher than the male. This is not the case with the Maoris measured by me.

The average index of the skulls in the Paris Museum is 96, while that of Professor Flower's series is 93·7, both considerably higher than what I give above. Professor Turner's have 88 as their average.

Combining their results with mine, we get, from a series of sixty-three skulls, an average orbital index of 90·1, a little higher than what I get from my own measurements, which brings the Moriori into the megaseme group.

The average interorbital distance in my skulls is between 20mm. and 21mm.

The group of male skulls is leptorhine, as also is the general average. The female skulls and the skulls of doubtful sex are low down in the mesorhine group.

The range of variation is 13·3.

Twenty-seven skulls, or 64·3 per cent. are leptorhine; fifteen, or 35·7 per cent., are mesorhine; none are platyrhine. The anterior nasal opening is therefore narrower in these Moriori skulls than in the Maoris that I have examined.

Professor Flower's eight skulls have an average index of 46·1; Professor Turner's eight give an average of 47·4; while that of the five in the Paris Museum is 47·5. The average when these four series are combined with mine—sixty-three skulls—is 46·8, the same as that given by my collection alone.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The above table of averages shows that all the divisions are orthognathous, though close to the upper limit of the group.

The range of variation is 11.

Nineteen skulls, or 46·3 per cent., are orthognathous. Twenty-two, or 53·7 per cent., are mesognathous. Seven have indices over 100, but none are actually in the prognathous group.

Seven of Sir William Flower's series allowed of the calculation of this index, and these yield an average of 99·2, some what higher than that of my set; while Professor Turner's have a mean index of 96, a trifle lower. The average of the whole fifty-six skulls is 97·6.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The range of variation is 10.

The projection of the nasal portion of the face, as shown by this angle, is greater in the Moriori than in the Maori skulls.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The average of each group is brachyuranic.

The range of variation is 22·6.

All the skulls with the exception of four are brachyuranic. The four exceptions are mesuranic. Exepressing this in percentage form, 89·7 per cent. belong to the former, 10·3 per cent. to the latter group.

A palato-maxillary index is not given in the catalogue of the College of Surgeons' Museum, nor the measurements by which it is calculated. I cannot therefore compare my skulls with those in that museum in this respect. Those measured by Professor Turner seem to have had narrower palates than those included in my table, as their average index is only 113. His range of variation, between 120 and 107, differs also considerably from mine. Combining his results with mine, we get a general average of 119·5 for 47 skulls.

The width of the face compared with that of the anterior part of the brain-case is great. In every adult skull, whether male or female, the zygomatic arches are visible from above. In eight, the arches, though visible, do not show their opening, but in the other thirty-four they project freely beyond the part of the brain-case above them. The maximum transverse diameter of the brain-case is, however, exceeded by the bizygomatic in only six skulls, while it is the longer of the two in thirty-one. In two these diameters are equal. This proportion is much the same as that noted in connection with the Maori skulls.

Basi-nasal Length or Cranio-facial Axis.—In thirty-two

males this averages 103·8mm. This is almost identical with what I find in the male Maori skull. The seven females have a mean diameter of 98·6. The extent of variation is from 116mm. to 98mm. in the males, while in one of the females the axis is 95mm.

In this dimension the Moriori skull is, like the Maori, exceeded in Professor Flower's table already referred to only by the Esquimaux, and to a very slight extent indeed by the Fijian.

Comparing the basi-nasal length with the rest of the mesial vertical circumference, as we did with the Maori skulls, we find that, taking the mean of all the measurements, the direct length from basion to nasion is 24·7 per cent. of the curve over the vertex between the same two points. This is the mean of both sexes combined. Or, making the similar comparison suggested by Professor Cleland, we get the proportion between the opistho-nasal diameter and the opistho-nasal arc to be as 1 to 2·69.

Treating the upper and lower divisions of the transverse vertical circumference of the brain-case in the same way we find that the mean infra-auricular distance is to the supra-auricular arc as 46·5 to 100.

Nasio-opisthic Median Arc.—The frontal portion of this curve exceeds the parietal and occipital in twenty of the thirty-two male skulls examined, while in two more, though longer than the occipital, it is equalled by the parietal. In five cases the parietal is the longest; in five also the occipital exceeds the other two. The parietal is the shortest in thirteen male skulls, the occipital in sixteen. The female skulls have the frontal arc longest thrice, the parietal also thrice, and the occipital once. The only case amongst either Morioris or Maoris in which the frontal is the shortest subdivision of the arc is the Moriori of doubtful sex, numbered 40 in Table III.

Shape of the Cranial Vault.—As already. mentioned, the frontal region in a large proportion of the skulls examined is in a very marked degree flat and retreating. This has been noted by other observers. The roof-like form of the vertex in the parietal region is also a very noticeable feature of the brain-case. The median ridge varies in height, but it is present in almost all skulls, and the parietal bone between it and the eminences is almost invariably either flat or even hollowed. Those few skulls which have more rounded vaults are, for the most part, female, or scarcely full grown. This type of vertex, especially when combined with the usually prominent parietal eminences, gives to most of these skulls, when viewed from behind, a markedly pentagonal form. The narrowness of the frontal region as compared with the parietal

has been already noted, and, in consequence, the skull as seen from above is decidedly obovate. The level of the maximum transverse diameter varies. In six skulls it is at the parietal eminences, in nine on the parietal bones lower down, in twenty-two on the squamous suture, and in four on the squamosal portion of the temporal bone. The temporal fossæ are large, and include the parietal eminences, and a post-bregmatic hollowing is not uncommon. The form and size of the glabella and superciliary ridges has been already described.

Sutures.—The coronal and sagittal sutures are, as a rule, simple. The lambdoid is complicated, sometimes exceedingly so. In many skulls the coronal suture is quite unserrated, except for a short distance above the stephanion. The part below the stephanion is always simple, and is frequently ossified; indeed, this suture shows partial or complete fusion much more frequently than does the lambdoidal. In those skulls which show commencing obliteration of the sagittal suture I have noted that the process had begun anteriorly in eight skulls; posteriorly, or in the region of the obelion, in six.

No skull in my series is metopic, but three show a short fissure in the glabella.

An interparietal bone is present in one skull—an adult male. In it a distinct serrated suture passes across the occipital bone from lateral angle to lateral angle. In three others, though the division of the occipital is not complete, yet there are fissures extending in towards the protuberance from the lateral angles. An interparietal bone was also present in one of the skulls examined by Professor Turner.

Pterion.—In the great majority of cases this is H-shaped, the length of the articulation between the parietal and the great wing of the sphenoid varying from 1mm. to 19mm., and averaging between 8mm. and 9mm. A K-shaped pterion was noted once. In another case the squamosal bone articulates with the frontal, but this is by means of a process projecting forward from the former bone, an epipteric which has fused with it. Free epipteric bones were noted twenty-six times in nineteen skulls. Sometimes these completely separate the four bones of the region, but in other cases the sphenoid and parietal articulate with each other either in front of or behind the epipteric bone.

Wormian Bones.—These are very often present, in addition to the epipterics just described. I have noted their occurrence in the lambdoid suture in twenty-eight skulls, in one completely separating the occipital from the parietal bones except for a very slight contact close to the asterion; at the asterion in ten; in the occipito-mastoid suture in eight; in the parieto-squamous in two. Small epactal bones were observed twice.

In the parieto-mastoid suture I have seen a wormian once; in the parieto-frontal, above the stephanion, twice; while in the suture, between the great wing of the sphenoid and the malar in the outer wall of the orbit, I have met with wormian bones in two skulls.

Paramastoid Process.—Two skulls show this projection on both sides. Two have a single process. One of these last, though broken, is 10mm. long.

Auditory Meatus.—In six skulls the lumen of the meatus is diminished, or almost obliterated, by a hyperostosed condition of its bony walls. In another, distinct exostoses are present. Two skulls show perforation of the tympanic plate. In one of these the meatus is much dilated, as if by a polypus or some other morbid growth. An exostosed condition of the meatus has been noticed by others as being not uncommon among the South Sea Islanders.

Parietal Foramen.—This is absent in eight skulls, but in several others it is very small. In twenty-three skulls it is present on one side, while in three there is a foramen in the line of the sagittal suture.

Posterior Condyloid Foramen.—The complete absence of this foramen was noted eight times. It is present on the right side alone in fourteen skulls, on the left alone in four.

Superior Curved Line of Occipital Bone.—This is, as usual, best marked in the male skulls, forming in one of them a very projecting ridge. In a large number of skulls it is distinctly double.

Equilibrium of Skull.—Posterior mastoid in eleven cases; anterior mastoid in one; double mastoid in four; posterior condyloid in two; anterior condyloid in four; posterior condylo-mastoid in two. Eighteen skulls rested on the teeth in front, and either on the conceptacula cerebelli or the posterior margin of the foramen magnum behind. In one the posterior point of support was as far back as the inion, while in another it was a paramastoid process.

Spheno-pterygoid Foramen.—In no case is this foramen present in its fully-completed form, but it is nearly perfect on one side in six skulls, and in another on both. In a somewhat less complete condition it is present on both sides in an eighth skull; whilst in ten more a bony spicule projects back-wards from the posterior margin of each external pterygoid plate towards the spine of the sphenoid. In all these cases the pterygoid plate is generally broader than usual.

Nasal Bones.—In some specimens the nasal bones are short, and the root of the nose is sunken; but more often they are long, narrow, prominent and convex below, and the root of the nose is not depressed, or only slightly so: in some it is even remarkably high. The Jewish nose, which is said to

have been frequently seen among the Morioris, should be remembered in this connection. It must not be supposed, however, that this form of nose is peculiar to the Morioris among Pacific Islanders. Such noses are not uncommon among the Maoris of some districts of New Zealand, and Professor Moseley^* states that they occur to the extent of 5 or 6 per cent. among the Admiralty Islanders.

Lower Margin Anterior Nares.—In twenty skulls this is rounded; in thirteen sharp, as in Europeans; in six the margin is doubled; while in four it is bevelled off.

Infra-orbital Suture.—In a large number of skulls this suture is present, either complete or partially obliterated. It was noted as present on both sides in twenty-one skulls.

Third Condyle.—Though no true example of this condition was observed, in one skull a smooth articular surface was present on the anterior margin of the foramen magnum. This was probably caused by the rubbing of the tip of the odontoid process.

Lachrymal Bone.—Unfortunately, owing to the broken state of the inner wall of the orbit in many of the skulls, I was unable to determine the nature of the articulations in this region in more than a few cases. In some of these, the only peculiarity is a marked shortening of the ethmo-lachrymal suture. In one this is reduced to 5mm., in another to 4mm., and in another to 1 ½mm. But six skulls show peculiarities which, I think, merit special notice. In the first, the lachrymal bone of the right orbit is a narrow, delicate, curved plate, separated from the frontal by a narrow neck of bone connecting the os planum with a backwardly prolonged nasal process of the superior maxillary. In front of and behind the lower part of this rudimentary lachrymal are spaces which were probably at one time filled by additional ossicles. Behind this the orbital process of the superior maxilla sends a triangular process upwards to within 2mm. of the frontal bone. In the left orbit of the same skull the lachrymal is still further reduced, and a similar process of the superior maxilla passes up between the nasal process and the os planum, and articulates with the frontal bone. In another skull (left orbit) there is no separate lachrymal bone, but its place is taken by the extension backwards of the nasal process of the superior maxilla, behind which a broad triangular process of the frontal passes down, completely separating it from the os planum, and articulating with the orbital surface of the superior maxilla. My collection, therefore, shows two examples of the rare orbito-maxillary-frontal suture. This articulation has been described and figured by Sir William

Turner^* and Mr. Arthur Thomson ^†, and it is worth noting that, though this condition was found only thrice in the 1,037 skulls in the Oxford collection examined by this last observer, yet one of these three skulls was from the Chatham Islands, while another was a Maori. The third was an Andamanese; and Professor Turner's specimens were bushmen. In another skull in my collection, in the left orbit, a triangular process from the frontal bone and one from the orbital plate of the superior maxilla approach each other between the os planum and lachrymal to within a millimetre. This small interval is filled by an ossicle distinct from both lachrymal and ethmoid. In the right orbit of the same skull the ethmo-lachrymal suture is only 1 ½mm. in length. In two other skulls—right orbit in one, left orbit in the other—the lachrymal bone is absent, its place being taken by a backward prolongation of the nasal process of the superior maxilla, which articulates with the os planum. A somewhat similar condition occurs in the right orbit of another skull, but two small ossicles are present in this maxillo-ethmoid suture, which almost completely separate the bones from each other. In this same orbit the frontal bone articulates with the orbital process of the palate behind the os planum. In the left orbit of this skull there is another modification of the same condition. The broadened nasal process of the superior maxilla articulates with the ethmoid behind; but the articulation is shortened below by the presence of an ossicle, which articulates with the orbital plate of the superior maxilla.

In the right orbit of another skull the osplanum is divided into three parts.

Alveolar Arch.—The curve is either parabolic or hyperbolic in almost all the specimens examined; but in three the molar portions are parallel to each other, and the arch would be classed by Broca as hypsiloid. In one skull there is a slight incurvation of the premolar region.

Lower Jaw.—The mandible is present in fourteen skulls, and shows considerable variation as to the degree of rounding of the angle and the character of the lower margin of the body, some bones being of the ordinary European type in these respects, others being convex from behind forwards, as we have seen is frequently the case with this bone among the Maoris. The coronoid exceeds the condyloid height in every case except one, and in it the two diameters are equal. The bigonial width also exceeds the gonio-symphysial length in every case. In one, however, these measurements differ by only 1mm.

Not only does the table show that the index of each vertebra is higher in the Maori, but also that, while in the European only one vertebra—the first—has the posterior surface of its body distinctly deeper than the anterior, in the Maori, on the contrary, there are three—the first, the second, and the third—deeper posteriorly than anteriorly.

In Sir William Turner's Maori the first and second vertebræ are also deeper behind than in front, the third and fourth are of equal depth on both surfaces, and the fifth is deeper in front than behind.

The following table shows to what extent the individual vertebræ agree with the average.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Looking now to the sexual differences as shown by the measurements, we find the average general index of the seven males to be 106·2, while that of the five females is 104·4. The relatively greater posterior depth is therefore shown by the lumbar vertebral bodies in both sexes, but the females have it in a less decided degree than the males, though, as before noted, the lowest index in the whole series belongs to a male column.

The table below gives the average indices of the individual vertebræ grouped according to sex, and shows that though both sexes clearly have their lumbar vertebræ built on the savage type, yet that those of the Maori woman, like her European sister's, are shaped in a manner more favourable to the lumbar curve than those of the male.

Breadth of entire pelvis: The maximum distance between the outer lips of the iliac crests.

Breadth between anterior superior iliac spines: The distance between the most prominent parts these eminences.

Breadth between posterior superior iliac spines: The distance between the inner borders of the spines.

Height of entire pelvis: From the highest part of the iliac crest to the most distant part of the tuberosity of the ischium.

Breadth between the outer surfaces of the tuberosities of the ischia: The maximum width.

Breadth between the ischial spines: Between their tips.

Breadth of Sacrum: The maximum width across the base.

Length of Sacrum: The distance in a straight line from the middle of the anterior border of the upper surface of the first sacral vertebra to the corresponding point on the lower border of the fifth. Even in sacra with six vertebral elements the same points are chosen.

Length of sacral arc: The length of the sacrum taken in a curve along the front of the bodies. By comparing this length with that taken in a direct line, the arc with its cord, a good idea is got of the degree of curvature of the bone.

Pubo-ischiatic depth: The distance between the upper surface of the pubis immediately in front of the ilio-pectineal eminence and the lowest part of the tuberosity of the ischium. This gives the depth of the pelvic cavity laterally.

Depth of pubic symphysis.

Transverse diameter of pelvic brim: The maximum width between the ilio-pectineal lines.

Sagittal or conjugate diameter of pelvic brim: From the middle of the upper margin of the anterior surface of the first sacral vertebra to the nearest point on the posterior surface of the pubic symphysis. When a median ridge is present on the pelvic face of the symphysis the measurement is taken somewhat to one side of the middle line.

Transverse diameter of pelvic outlet: Taken as recommended by Dr. Garson, between the most widely separated points on the lines which pass forwards from the ischial spines towards the lower ends of the obturator foramina.

Sagittal diameter of pelvic outlet: From the centre of the lower margin of the fifth sacral vertebra to the lowest part of the pubic symphysis.

Subpubic angle.

Height of acetabulum: From the upper margin below the anterior inferior iliac spine to the opposite border of the cavity.

Breadth of acetabulum: Taken across the cavity at right angles to the previous measurement.

Height of obturator foramen.

Breadth of obturator foramen.

Interobturator width: The distance between the inner margins of the obturator foramina.

Breadth of innominate bone: The distance between the angle of the os pubis and the most distant part of the iliac crest, generally a little above the posterior superior iliac spine.

Breadth of ilium: The distance between the anterior superior and posterior superior spines of the ilium.

As far as possible, the measurements about the innominate bone were taken on the right side.

The indices used, and the formulæ by which they are calculated, are as follow:—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The following table gives the breadth and height of the whole of the pelvis, and the height-breadth index, or proportion between the two, distinguishing the sexes, and giving the maximum and minimum as well as the average in each case.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The greater height of the male pelvis, not only absolute, but relative to width, is clearly shown in the table.

The measurements of two Maori pelves, both males, are given by Sir William Turner.^* The maximum width of one is 255mm., and its height is 221mm.—a pelvis, therefore, somewhat narrower and higher than the average of my male specimens. Its index is 87, higher than that of any in my series. In the other pelvis, which was scarcely full grown, the breadth is 238mm., the height 188mm., and the index 79mm.

These twelve male pelves are, on the whole, both narrower and shallower than the average pelvis of the European male, but the height is proportionally to the breadth slightly greater in the Maori, as is seen by comparing the height-breadth index in the two races. The average male index, as given by M. Verneau, in Europeans is 79.

The female Maori pelvis, so far as shown by the fourteen now described, is also narrower than that of the European female, but is its equal as to height. It has therefore a higher index, that of the European female being 74 according to M. Verneau, 75 according to Dr. Garson.

In both sexes, then, but to a slightly greater extent in the female than in the male, the Maori pelvis is higher relatively to its width than is the case with the majority of Europeans.

The following table, arranged like the last, gives the average measurements and index of the brim of the pelvis:—

The figures show that while the two diameters of the brim are, as is always the case, longer in the female than in the male, yet that the proportion between them is much the same in the two sexes, the female showing a slightly greater proportional length in the sagittal diameter.

It is curious that both the highest and the lowest index should be found among the female pelves. The maximum index, 102·3, does not, however, belong to a pelvis with an unusually short transverse diameter. On the contrary, this is 128mm., a little above the female average. It is the great length of the antero-posterior measurement which is the cause of this exceptionally high index. This is the only pelvis of either sex in which the antero-posterior diameter of the brim exceeds the transverse.

Professor Turner's two male pelves have higher brim indices than of any of my males. In one the index is 97, and in the other it is 95. Even allowing for the fact that Professor Turner measures to the back of the upper end of the symphysis, the indices show that in these two pelves the conjugate diameter is longer proportionally to the transverse than in any of the male pelves that I have measured. Including these two pelves with mine, we get an average

brim index for males of 87·9, a little higher than that given in the table, and almost the same as that yielded by the measurements of the female pelves.

If we now compare the Maori and European pelves with regard to these measurements of the brim, we find that the antero-posterior is longer relatively to the transverse in the Maori of both sexes than in the European. M. Verneau gives the brim index in European males as 80, in females as 78; Professor Turner gives 77 for males and 79 for females as the result of his measurements; Dr. Garson gives 80 for females; Sir William Flower, 81 for males, 78 for females; and, though some anatomists give higher results, those of others—as, for example, Martin, who gives 69 as the brim index of Irish women—are lower.

We may safely say, then, that the Maori pelvis, with its index of nearly 88, is narrow compared with the European type, but it falls into the same group if we follow Sir William Turner's classification. He divides pelves into three groups—platypellic, those with a brim index below 90; mesatipellic, those whose index is from 90 to 95; and dolichopellic, those whose index is above 95. Europeans, Chinese, and some savage races, such as American Indians and Fuegians, belong to the first group; Negroes are found in the second; while Australians and Andaman Islanders are members of the third. According to my measurements, the Maoris, both male and female, are also platypellic. All the pelves in my series, however, do not belong to this group. Fifteen—seven males and eight females—do so; but six—three males and three females, are mesatipellic; and the remaining three females are dolichopellic. Professor Turner's two males are also dolichopellic.

The depth of the pelvic cavity, as shown by the puboischiatic diameter, and its relation to the transverse diameter of the brim, are shown in the following table.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The difference between the average depth in the male and in the female is seen to be 6·7mm. In a previous table it was shown that the average height of the entire male pelvis exceeds that of the female by 13mm. The lower height of the

Comparing these figures with the transverse diameter of the pelvic brim, we find that in the male pelvis the sacral width is 97 per cent. of the brim width, while in the female it is 94 per cent., showing that the curvature of the back part of the iliopectineal lines is more pronounced in the female than in the male.

It will also be seen that the average degree of curvature of the sacrum, as got by comparing the length with the are of the bone, is practically the same in both sexes.

Though the female index is distinctly higher than the male, it is worth noting that the highest and lowest indices both belong to female sacra.

The length of the sacrum in Professor Turner's two male Maoris is 101mm. and 120mm., while the breadth is 97mm. and 115mm. The sacral index is in each case, therefore, 96. If we include these with my set of males we get an average for that sex of 106·5, nearly eleven less than the average female index.

According to M. Verneau's measurements, the average length of the European sacrum is 105mm. in males, and 101mm. in females, while the width in the male is 118mm. and in the female 116; though Dr. Garson gives this latter average as 118·3. The Maori sacrum, then, so far as my measurements go, is in both sexes of almost the same length as the European; but it is somewhat narrower in the male and broader in the female than the European average for each sex.

Looking to the proportion between these two measurements, the index, we see that the Maori sacrum is in both sexes broader than it is long, and, in consequence, belongs to the platyhieric group (platyhieric, index over 100; dolichohieric, index under 100)^*, in common with Europeans. The male sacra are, however, almost in what Professor Paterson^† calls the subplatyhieric division—those with an index between 100 and 106, a group containing, amongst others, Andaman Islanders and Negroes. The female sacra have, on the other hand, indices equal to the average of the European female. Only one sacrum is dolichohieric, and that is a female with an index of 99·2.

As five of my pelves belong to skeletons of which I have also the skulls, I give in the following table some of the more important indices of the two cavities in each body.

sured by him as 64·7 and 73·5. A well-marked degree of platycnemia is therefore characteristeric of the Maori tibia, contrasting strongly with what the European bone shows in this respect, to which an index of 73·7 is given by Thomson as the result of the measurement of twenty-one specimens. But even among races of a corresponding civilisation the shaft of the Maori tibia is more than usually flattened, as may be seen by referring to the tables of this index as it occurs among other savage races, given in M. Topinard's book and in Mr. Thomson's papers on the subject.

I have also, following Mr. Thomson, examined all the bones that have come under my observation as to the anteroposterior curvature of the external condyloid surface. I have noted it as convex in twenty-eight cases, as plano-convex from before backwards in ten, as flat in one, and as concave in five.

Mr. Thomson has also been the first to note the presence of a facet on the anterior margin of the lower end of the tibia. This, though occasionally met with in Europeans, is of much more common occurrence among savages, and is caused by the pressure of the upper surface of the neck of the astragalus. This condition I carefully looked for, and in every bone examined except two a greater or less degree of flattening or hollowing was found to be present. The antero-posterior curve of the shaft is also more pronounced than among Europeans; but, while in some bones the curve is fairly regularly distributed along the length of the bone, in others it is mainly confined to the upper fourth.

That the squatting posture is the cause of these last peculiarities admits, I think, of no doubt. They are met with in the living races that adopt that attitude, and are all readily explained by the contacts, pressures, and muscular strains consequent on that position, which was also universal among the Maoris at the time when the individuals lived whose skeletons I have examined. Till recently, however, platycnemia has not been put down to the same cause. It is generally said to be due to the muscular development—tibialis posticus—caused by an active life in a rough, hilly country. Professor Havelock Charles,^* however, shows that the tibia of the natives of the Panjab have not only the altered outer condyle, the curved shaft, and the anterior astragaloid facet, but has an average index of platycnemia of 69·9. A mountainous country can have nothing to do with the flattening in this case, while squatting is the universal custom. And I think it not impro-

ing that we have seen to be characteristic of the Maori femur, and almost to the same extent. Platycnemia is slightly more pronounced than in the Maoris, and the tibia shows all the other characteristics due to the squatting posture which have been already described.

The description of the Maori skull contained in the preceding pages agrees in all essentials with that already given by other observers. It is, according to my measurements, mesaticephalic, though on the verge of dolichocephaly; metriocephalic; mesoseme; mesorhine, though almost leptorhine; orthognathous; brachyuranic; phænozygous: and the males are megacephalic.

Though the Maori cranium has, so far as its race characters go, been exhaustively studied—nearly two hundred skulls having now been more or less fully described—few have had the opportunity of examining the other bones of the skeleton. I have been fortunate in this respect, and have been able to show that the vertebral column is typically savage in the form of its component parts; that the pelvis is platypellic, with a platyhieric, almost subplatyhieric, sacrum; that the upper limb is mesatikerkic; while the lower limb, though brachycnemic, is on the verge of the dolichocnemic group; that platycnemia is well marked; and that the limb-bones generally show the modifications of form characteristic of most of the coloured races.

If any further proof were needed of the mixed origin of the Maori race it is given in this paper. An examination of the cranial indices and of the extent of their variation shows this clearly. These demonstrate two distinct types and intermediate forms. At the one extreme we have skulls approaching the Melanesian form, as met with in the Fiji group, long and narrow, high in proportion to their breadth, prognathous, and with wide nasal openings. At the other are skulls of the Polynesian type, such as are common in Tonga and Samoa, shorter and broader, with orthognathous faces. And it must be noted that these extreme forms do not belong to different tribes or districts, but may both be found in one. Among the skulls of the Ngaitahu Tribe alone we have as great a variation in almost all the indices as is met with in the entire collection of crania gathered together from all parts of both these Islands.

Though a thorough mingling of the two parent stocks has thus taken place, it is yet clear that their distinguishing characters are not seen in the same proportion in each district. Undoubtedly intertribal differences exist. The material at my disposal has allowed of nothing more than a comparison

between the crania of the Ngaitahu Tribe of the South Island and those from the northern part of the Auckland Province, mainly from the Bay of Islands and Whangarei. The Ngapuhi are now the most important tribe in this region, but I cannot say whether the skulls included in this group belong to this or, indeed, to any one tribe. The district, however, is clearly defined, and the skulls available closely approximate in number to the Ngaitahu examined by me. By availing myself of the twenty-two crania already described by Professors Flower and Turner, I get, with those whose measurements are now given, a set of thirty-four skulls to compare with the forty of the southern tribe. The other two groups from the North Island given in the paper are too small to be of any value in estimating tribal characteristics, and I institute in this paper no comparison between them and the two larger collections. In the Auckland skulls the average cephalic index is 73·3, the vertical 72·8, the frontal 68·2, the orbital 89·3, the nasal 50·2, the gnathic 98·4, the palato-maxillary 119·9; while the same indices in the Ngaitahu are 75·9, 74·6, 67·2, 86·4, 48·0, 97·6, and 120·8; showing that the skulls of the former group are, proportionally to their length, longer than those of the latter, that they have wider nasal openings, and slightly more projecting jaws. And it has also been shown that the proportion of height to breadth is slightly greater in them than in the Ngaitahu. The Melanesian characters are therefore more accentuated in the North than amongst the Natives of the South Island. The prevalence of the Papuan form among skulls from the Bay of Islands has also been observed by MM. de Quatrefages and Hamy, and is noted by them in the “Crania Ethnica”^* when describing the Maori cranium.

The measurements now given of the Moriori skull, taken with those already published, show it to be mesaticephalic, though close to the lower limit of the group; metriocephalic, though almost tapeinocephalic; low down in the megaseme group; leptorhine; orthognathous; brachyuranic; phænozygous: and the males to be megacephalic.

It differs from the Maori skull mainly in its lesser height, both absolute and relative to length and breadth; the greater excess of the parietal over the frontal width; the higher orbits; and the narrower nasal opening. The depressed and retreating forehead is also a very marked feature of many Moriori skulls. It is slightly broader relatively to its length, and somewhat more prognathous. The cranial capacity is also slightly less. But, as already pointed out, there is often a very close resemblance between Maori and Moriori skulls.

The variation of the indices, though somewhat more restricted than with the Maoris, is still considerable, and points, like the traditions of the people, to an origin from the two great Pacific stocks. The different types of Moriori skull have been already sufficiently described.

Though the number of skulls available for measurement is sufficiently large, I regret to say that I have been unable to secure more than five Moriori skeletons, none of which are quite complete. As nothing definite as to the osteological characters of the race can be deduced from such a small collection, I have done little more than give the measurements and indices, with their averages. These show in most instances a very close correspondence with what we find among the Maoris.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]