to be the moulding of an African people (all trace of the aboriginal Arawak and Spaniard being completely lost in the all-prevailing negroid type) by English and Scotch life and thought, no other influence having come in contact with the race. Here, too, it appears possible to have one civilisation alike for black, coloured, and white, which is not the case in America. The reasons given for this rapid evolutionary development since the emancipation of the slaves in 1834-38 are: (1) the security of a solid government; (2) widespread education afforded by 757 schools; (3) an active religious propaganda rapidly suppressing Obeahism; (4) easy conditions of life: the scant needs of the negro are easily met; (5) stateaided settlement of lands on deferred payments, which is establishing a growing class of peasant-proprietors.

3. Mongoloid Europeans. By DAVID MACRITCHIE.

A careful consideration of the relics of the Cave-men of Europe has led Professor Boyd Dawkins to the conclusion that the Eskimos of the present day are almost certainly their representatives, and that the connection between these two peoples must be one of blood. He does not ignore the possibility of descendants of the Cave-men having survived into historic times in Europe; but he is of opinion that the Eskimo type has long been extinct in Europe.

Here he is at variance with the deductions of Dr. Beddoe, made after an analysis of the race-elements in modern Britain, to the effect that some reason can be shown for suspecting the existence of traces of some Mongoloid race in the modern population of Wales and the west of England.'

While stating his belief that the Cave-men of Eurasia were driven eastward into North America, where their descendants now exist as Eskimos, Professor Boyd Dawkins points out that North-eastern Siberia yet retains an Eskimo population-the Chukches. Martinière, however, reports in the Yalmal peninsula, in 1653, a people closely resembling modern Eskimos in physical appearance, in dress and manners, and, above all, in their use of the peculiar skin-covered skiff generally known as a kayak. These skin-canoes are not reported in Arctic Europe during the last few centuries; but they are said to have been used in the Orkney Islands by a race of occasional visitors, locally called 'Finnmen,' between the years 1682 and 1701. The minute accounts given of the canoes of these Finnmen leave no doubt that they were kayaks. One of them was preserved in the Church of Burra, Orkney, for a time; and another in the Physicians' Hall, Edinburgh. Popular tradition in Orkney and Shetland contains many references to these Finnmen' or 'Finns,' who are said to have frequently intermarried with the islanders. These islanders are mainly Norsemen, and in the Norse language 'Finn' signified'Lapp.' The historical statements and the traditions relating to 'Finns' denote really, therefore, an intercourse and a partial fusion between the islanders and a Lapp people accustomed, like modern Eskimos, to the use of the kayak and all that that implies. Such a fusion would readily explain the Mongoloid features seen in certain Shetlanders by Dr. Beddoe.

It is noteworthy that the territory occupied by the Lapps in the ninth century included the greater part of Scandinavia, and straggling remnants of that population may have survived for many centuries in Southern Scandinavia. Martinière even speaks of a Lapp village near Christiania in 1653.

There seems to be no trace of the use of the skin canoe among modern Lapps, but von Düben states that the mountain Lapps assert that their remote forefathers, who came from the south-east of Europe, crossed the sound which separates Denmark from Sweden in small skin-boats.

If this tradition be accepted as accurate, it is reasonable to suppose that remnants of the Lapps who were able to prolong their existence among the fiords, long after the days of Norse' invasion from the Continent, would at the same time continue to use the skin canoes of their race, and this long after the inland Lapps had ceased to know anything of such vessels, except from tradition. This hypothesis would readily account for the existence of coast-dwelling Lapps who

crossed from Scandinavia to Orkney and Shetland in their kayaks as recently as the seventeenth century.

This conclusion supports Dr. Beddoe's belief that there is a Mongoloid element among western Europeans. If the Mongoloid people assumed by Professor Boyd Dawkins to have existed in Western Europe in primitive times really died out, it seems necessary to suppose that there was a fresh Mongoloid immigration at a much later date, e.g. the Hun conquests of the fifth century. But there appears to be ample evidence that Europe contained a truly Mongoloid population long before the era of Hun domination, and even that the European Cave-men have never ceased to be represented by people who have inherited their blood.

4. Some Points about Crosses, chiefly Celtic. By Miss A. A. BULLEY.

The paper deals with certain details only, and has nothing to do with the general question of origin. In considering the form of the crosses, however, regard is had to the feeling underlying the treatment, so far as this can be gathered from a general survey of the examples. Argument from form alone is necessarily imperfect, and may be fallacious, though it may suggest lines of investigation. In default of historic data, however, it is the only method possible.

1. Celtic crosses.-From a survey of examples from Cornwall, Wales, the Isle of Man, Scotland, and Ireland the author infers that in Celtic crosses (i.) the circle (whatever its meaning and however related eventually to the ends of the arms) is not a mere adjunct (such as a glory, or a support for the arms), but is of at least equal importance with the cross. The persistence of such a form without meaning points to an earlier period when the form represented an idea of primary importance. The circle is therefore inferred to be here a root-idea. (ii.) The long-shafted or Latin type appears to be an independent development from the cross-with-circle-and-equal-arms. The author does not attempt to decide whether or no this development was influenced by the introduction of the pure 'Latin cross' from outside.

2. Non-Celtic crosses, on the other hand, exhibit lesser importance and weaker treatment of the circle, e.g. :

(a) Coptic crosses, though often inclosed in a wreath, are often without. Later, the ankh symbol is confused with the cross.

(b) Roman (catacomb) crosses in their earliest form are equal-armed, but the circle is optional. The long-shafted or Latin' form is later, and possibly developed (in Italy) from processional use. The treatment also of these early crosses is not realistic but symbolic; whereas the course of development is never from symbolic to realistic, but the reverse.

(e) Syrian crosses resemble catacomb-crosses, but are even more decorative in treatment. The extant examples, however, are chiefly architectural ornaments, which may account for this. The circle is optional.

5. Some Suggestions as to the Origin of the Brooch, and the probable Use of certain Rings at present called Armlets.' By EDWARD LOVETT.

The author suggests, as the prototype of the ring-and-pin contrivance for fastening a cloak, the use, by a hunting people, of the mammalian Os innominatum and Os calcis, the corners of the cloak being drawn through the oval perforation of the former and then pierced by the sharp point of the latter. In this position the prominences on the Os calcis would drop into the hollow of the Os innominatum and prevent the Os calcis from falling out of place.

The author notes, further, that very many rings of early date and various materials-bone, jet, shell, bronze, and iron-which are usually described as ' armlets' are of too small diameter to allow the entrance even of an infant's hand. As such rings are frequently found associated with pins of similar materials,

commonly regarded as 'hair-pins,' and as ring and pin are sometimes found in situ on the breast of a skeleton, it is inferred that they represent a simple ringand-pin fastening of the kind described above. An apron-fastener of this type, composed of an iron ring and a horse-shoe nail, is still worn in some of the blacksmiths' shops in Scotland.

The next step of development is taken when the pin is perforated at the thick end and attached to the ring by a fibre to prevent it from being lost. This stage is actually represented by a ring-and-pin fastening which is in common use in China: the ring is of agate, and the pin, which is of silver, is attached to it by a silken thread. Probably many of the perforated pins in our museums were similarly attached to rings.

An apron-fastener of the simple ring-and-pin type, composed of an iron ring and a horse-shoe nail, is still worn in some of the blacksmiths' shops in Scotland; a similar simple brooch is still worn by the shepherds of Perthshire and by the tinkers in this and other parts of Scotland; and another similar form was in very common use in Donegal as late as 1860.

A further step is taken when the pin itself is hinged upon the ring, for security, by bending its flattened head round the ring. This form is abundant in Celtic times. The Tara brooch is a striking example, though the author suggests that it may be a symbolic reversion to an earlier type.

The inconvenience which accompanies the use of the ring-and-pin brooch, that the fabric to be fastened must be drawn far through the ring before the pin can pierce it, was remedied, it is suggested, by leaving a gap in the ring; and from this results the 'penannular' brooch with its many varieties.

6. On the Ethnology of the Siciutl Indians of British Columbia,1
By C. HILL TOUT.

7. On the Canadian Indians as they are. By DAVID BOYLE.1

8. On the Legends of the Dieri and Kindred Tribes of Australia.1
By A. W. HoWITT and OTTO SIEBÈRT.

9. A West Indian Aboriginal Wooden Image. By J. E. DUERDEN, Ph.D. This figure represents one of the most characteristic types of West Indian wooden images, several of which are now known from different islands. They have been found mostly in caves, and historic references to such objects of worship or zemes in Columbian times are available. The present example represents a single crouching human figure, terminated above by a large circular canopy and resting upon an irregular wooden base. The face is very large; the ears are indicated by an upper smaller and a lower larger lobe, both perforated. The eye and mouth apertures are formed in the usual rounded manner, with thickened margins. The arms and legs are constricted, as by the wearing of circular bands; small mammæ, ribs, and a large erect virile organ are indicated.

10. On a Model of the Arbor Low Stone Circle. By H. BALFOUR, M.A. To be published in full in Journ. Anthr. Inst. xxxiv.

SECTION K.-BOTANY.

PRESIDENT OF THE SECTION.-A. C. SEWARD, M.A., F.R.S.

THURSDAY, SEPTEMBER 10.

The President delivered the following Address:

IN 1883, the date of the last meeting held by the British Association at Southport, the late Professor Williamson, of Manchester, delivered a Presidential Address before the Geological Section, in which he reviewed recent progress in palæobotanical research, with special reference to the vegetation of the Coal period. It would have been an interesting task to traverse the same ground to-day, in order to show what a vast superstructure has been built on the foundations which Williamson laid. In alluding to the controversies in which he bore so vigorous a part, Williamson spoke of the conflict as virtually over, though still reflected in the ground-swell of a stormy past.' Now that twenty years have elapsed we are able to recognise with no little satisfaction that his views are firmly established, and that the debt which we owe to his able interpretation of the relics of Palæozoic plant-life is universally acknowledged. Williamson's labours demonstrated the possibilities of microscopical methods in the investigation of Carboniferous plants; but at the time of publication his results did not receive that attention which their importance merited, and it is only in recent years that botanists have been induced to admit the necessity of extending their observations to the buried treasures of bygone ages. We have been slow to realise the truth of the following statement, which I quote from an able article on Darwinism in the Edinburgh Review' for October of last year: The recognition of the fact that in every detail the present is built on the past has invested the latter with a new title to respect, and given a fresh impulse to the study of its history.' The anatomical investigation of extinct types of vegetation has done more than any other branch of botanical science in guiding us along the paths of plant-evolution during the earlier periods of the earth's history.

I cannot conclude this brief reference to Williamson's work without an expression of gratitude for the help and encouragement with which he initiated me into the methods of palæobotanical research.

FLORAS OF THE PAST THEIR COMPOSITION AND DISTRIBUTION.

:

Introduction.

It is by no means easy to make choice of a subject for a presidential address. There is the possibility-theoretical rather than actual of a retrospective survey of modern developments in the botanical world, and the opportunity is a favourable one for passing in review recent progress in that department of the science

which appeals more especially to oneself. In place of adopting either of these alternatives, I decided to deal in some detail with a subject which, it must be frankly admitted, is too extensive to be adequately presented in a single address. My aim is to put before you one aspect of palæobotany which has not received its due share of attention: I mean the geographical distribution of the floras of the past. In grappling with this subject one lays oneself open to the charge of attempting the impossible-a not unusual characteristic of British Association addresses. I recognise the futility of expecting conclusions of fundamental importance from such an incomplete examination of the available evidence as I have been able to undertake; but a hasty sketch may serve to indicate the impressions likely to be conveyed by a more elaborate picture.

One difficulty that meets us at the outset in approaching the study of plant distribution is that of synonomy. 'The naturalist,' as Sir Joseph Hooker wrote in his Introductory Essay to the Flora of New Zealand,' 'has to seek truth amid errors of observation and judgment and the resulting chaos of synonymy which has been accumulated by thoughtless aspirants to the questionable honour of being the first to name a species." Endless confusion is caused by the use of different generic and specific names for plants that are in all probability identical, or at least very closely allied. Worthless fossils are frequently designated by a generic and specific title: an author lightly selects a new name for a miserable fragment of a fossil fern-frond without pausing to consider whether his record is worthy of acceptance at the hands of the botanical palæographer.

An enthusiastic specialist is apt to exaggerate the value of bis material, and to forget that lists of plants should be based on evidence that can be used with confidence in investigations involving a comparative treatment of the floras of the world. As Darwin said in the Origin of Species': 'It is notorious on what excessively slight differences many paleontologists have founded their species; and they do this the more readily if the specimens come from different sub-stageo of the same formation.' It would occupy too much time to refer to the various dangers that beset the path of the trustful student, who makes use of published lists of local floras in generalising on questions of geographical distribution during the different eras of the past. Such practices as the naming of undeterminable fragments of leaves or twigs, the frequent use of recent generic names for fossil specimens that afford no trustworthy clue as to affinity, belong to the class of offences that might be easily guarded against; there are, however, other obstacles that we cannot expect to remove, but which we can take pains to avoid. An author in naming a fossil plant may select one of several generic names, any of which might be used with equal propriety; individual preferences assert themselves above considerations as to the importance of a uniform nomenclature. The personal element often plays too prominent a part. To quote a sentence from a non-scientific writer: The child looks straight upon Nature as she is, while a man sees her reflected in a mirror, and his own figure can hardly help coming into the foreground.'

In endeavouring to take a comprehensive survey of the records of plant-life, we should aim at a wider view of the limits of species and look for evidence of close relationship rather than for slight differences, which might justify the adoption of a distinctive name. Our object, in short, is not only to reduce to a common language the diverse designations founded on personal idiosyncrasies, but to group closely allied forms under one central type. We must boldly class together plants that we believe to be nearly allied, and resist the undue influence of considerations based on supposed specific distinctions.

The imperfection of the Geological record was spoken of by one of England's greatest geologists, in a criticism of the Origin of Species,' as 'the inflated cushion on which you try to bolster up the defects of your hypothesis.' On the other hand, Darwin wrote in 1861: 'I find, to my astonishment and joy, that such good men as Ramsay, Jukes, Geikie, and one older worker, Lyell, do not think that I have in the least exaggerated the imperfection of the record. No one in the least familiar with the conditions under which relics of vegetation are likely to have been preserved can for a moment doubt the truth of Darwin's words: 'The crust