Extracts from his 1893 publication. This only full-length publication during his long career of more than fifty years at the Pitt Rivers Museum:

It is to General Pitt Rivers without doubt that we owe the stimulus which has of recent years led many workers to investigate the gradual development of the various Arts of mankind, and to endeavour to trace their histories back to their absolute origins. The illustration of the gradual growth of Decorative Art from simple beginnings was a part of his scheme for establishing series of objects with a view to tracing the stages in the evolution of all the material arts of mankind. With this object in view he accumulated material and formed series to illustrate the origin, growth step by step, and variations of certain patterns, and these series form some of the most interesting and striking features in his collection. Amongst many others one may mention the series illustrating the 'varieties and geographical distribution of the Loop Coil series showing the gradual degradation of designs representing the human form and their conversion into meaningless ornament, illustrating the importance of successive copying as a factor in the evolution of patterns; those showing the derivation of patterns on gourds and pottery from the strings by which once the vessels were carried. These and others such are too well known to require description. It is much to be regretted that with the exception of valuable remarks embodied in essays upon wider subjects, and a few descriptions of the development of special patterns, we have not the benefit of his researches into the history of Decorative Art in a published form. Among the earlier researches into the history of patterns should be mentioned Dr. John Evans's striking series of British coins, upon which, in the course of successive reproductions the once realistic design becomes hopelessly conventional (Proc. Royal Inst. vii. p. 475). Since the arrival of General Pitt Rivers's Collection in Oxford, as a gift to the University, I have constantly endeavoured to collect fresh material and facts, with a view to the further development of the subject. According to the general law laid down by him, and from the evidence now accumulated, there seem to me to be deducible certain general conclusions regarding the earlier history of Art, shedding light upon the first efforts of Prehistoric man in the Art of Design and Decoration, and also explaining the gradual formation and raison d'etre of various forms of ornament which are familiar to us. ...

[From page 17]

... We are justified, therefore, in appealing to the study of the primitive arts of savages, in order to elucidate points concerning which the evidence afforded by archaeology is incomplete. Assuming this I will pass on to consider the light which the art of modern savages throws upon the study of the origin and growth of Decorative Art. By the examination and comparative study of various patterns and designs, and especially by means of series illustrating the variations upon particular designs, we may arrive at certain conclusions as to how these have grown up from earlier stages, and in some cases trace their evolution back to their absolute origin as patterns. At the same time many of my illustrations will be drawn from the art of civilised people, in which the raison d'etre and evolution of patterns and designs are frequently very clear.


As in the useful arts so also in the fine arts we find the lowest (least cultured) savages deriving their early ideas from Nature. We find very much that in the matter of the Aesthetic Arts the condition corresponds with their primitive state of general culture. The ornamentation, for example, of their weapons and other implements is for the most part extremely simple, frequently, in fact, consisting of nothing more than natural peculiarities in the material, it may be knots in the wood, or the nodes on a reed stem, which are slightly emphasised with colour or otherwise in order to increase their decorative effect. General Pitt Rivers has given a good example of the adoption of a natural feature as ornament. In the collection formed by him there is an Australian 'boomerang' of unusually light-coloured wood, the curve, as usual, following the grain of the wood. Along the length of the weapon in the central line are a number of small natural knots in the grain, situated at fairly regular intervals. The peculiarity of this regular succession of knots evidently attracted the notice of the maker of the weapon, and, in order to increase the decorative effect, he stained each knot a dark colour, thus throwing these into greater prominence upon the light ground colour, and, moreover, to many of the dark patches he gave a more or less lozenge-shaped outline, in order to further enhance their value as ornament.

The decorative effect of the nodes upon a reed-stem or bamboo, due to the regularity of their arrangement, has appealed to the minds of savages in various parts of the world, Australia, New Guinea, Africa, and other parts, where we find spear- and arrow-shafts of reed, in which the nodes have been scraped smooth, and the bands thus formed round the stem have frequently been picked out in black, red, or white. ... From the mere appreciation of the uncommon to the artificial increasing of the effect, the step is, as I have said, but a slight one; but in a low state of culture, such as that of the less cultured savages, natural progress advances by extremely slow stages, never by sudden strides. The savage mind is receptive to a certain extent only, and is unable easily to grasp new ideas if they are greatly in advance of existing knowledge; they must be led up to by easy stages.

The savage peoples from whom I have drawn these examples of primitive ornamentation have, it is true, reached far beyond the earliest stages in the art of design; their art is no longer in its infancy. The art of the Australians, as of many savage races, is undoubtedly in a rudimentary state, but it nevertheless shows signs of steady progress during past ages, in the differentiation of its branches, and the skill sometimes displayed in the application of ornament. In savage art we find, as I have said, much evidence of its having been schooled by Nature, and the examples of Nature-suggested forms of ornament, such as those I have mentioned, seem to reflect the birth of the art and to be survivals of early phases in the history of its progress; they may, in fact, serve to suggest to us how the art of design first dawned in the remotest ages.

There can, I think, be little doubt that both graphic and plastic arts were in the first instance suggested to man in simple ways of this kind, and their origin should be referred back to the time when man's aesthetic appreciation of peculiarities, either natural or produced as accidents in manufacture, was sufficiently developed to suggest the application of artificial means in order to increase their effect; in other words, to control them to serve the special purpose either of representation of other objects or of ornament.

The first stage in the development of design as a fine art was purely what may be termed an ADAPTIVE stage, that is, man simply accepted and adapted effects which were accidentally suggested to him. ...


As a natural result of the appreciation and adaptation of natural or accidental effects, there arises a desire to produce artificially similar effects where these do not exist. For this a creative operation of the intellect is required, and it is here that the art develops fresh importance, and assumes definite vitality. Imitation is the mother of art, and is the outcome of this desire to possess some object or to reproduce some effect which is admired; it is inherent in our nature, and is perhaps the principal stimulus in the early development of the fine arts.

As a natural outcome of the development of the art of copying at first hand we come to successive copyings and the importance of this latter process in the development and progress of the art of design cannot be over-estimated. So long as the intention in each case is merely to make as accurate an imitation of the original as possible, it would seem that copying, whether successively or at first hand, should be considered as belonging to the Second Stage; but, as the effect of this successive copying is in all cases to create changes in the original design, even though unintentionally, and to render the primary conception very unstable, it is necessary to treat of this under the Third Stage, and to explain in connection with this stage the meaning which is here assigned to the expression 'successive' copying, and the value of this as a factor in the evolution of designs.


This, which is the resultant principally of successive copying, may be termed the stage of variation and must be divided into two sections:

(a) Unconscious Variation.

(b) Conscious Variation.

(a) Unconscious Variation—To a highly skilled artist it is no very difficult matter to make a copy of a simple object or pattern, which shall so resemble the original as to be hardly distinguishable from it. With people not specially trained, whether civilised beings or uncultured savages, and so we must believe also with primaeval man, it is different. In un- skilled hands and with indifferent tools accurate copying is an impossibility, and each new attempt at representing an object creates a variation from the original type.

Let us suppose that some one, whom I will call A, copies an object, and B copies A's version of it without having seen the original, and C copies B's, and so on; in each case the new copy varies from the immediately preceding one more or less according to the skill of the artist. We can readily see that in the course of time by such successive copyings designs can arise, which may entirely lose all resemblance to the original object, and to A's would-be realistic version of it.

It would be almost impossible to obtain from specimens of savage art a really consecutive series of any length to illustrate this point, as, in order that one may be sure of the absolute continuity of the line of succession, the series must be produced under circumstances which admit of constant supervision. It is, moreover, doubtful whether a long series of changes in a design is often, or ever, produced entirely by unconscious variation. Still, as this is so important a factor in the evolution of ornament and the conventionalisation of original realistic designs, it is necessary to have some illustration of the workings of this process, in order to show how rapid and complete may be the changes effected by it.

With this object in view I have formed some series after a method originated by General Pitt Rivers and suggested to me by him some while ago. Briefly thus: I first made a rough sketch of some object which could easily be recognised. Then I procured a number of pieces of paper of the same size as that on which the sketch was made. Next I enlisted the aid of a number of people who, while having some notion of copying designs, were not by any means skilled in the art (this in order not to make the series unnecessarily long, and in order to adhere to a certain extent to the conditions of primitive copying; to this end also the copies were made with a pen and not with pencil, as the latter, with its attendant possibilities of rubbing out, would have rendered greater accuracy possible).

Balfour 'Evolution of Decorative Art' Plate I

To the first, A, I gave my sketch, of which he made as accurate a copy as he was able on one of the slips of paper. I then withdrew my original, and set the second person, B, to copy A's version, which was then withdrawn; the third copied B's sketch; and so on; in every case all the former sketches were withdrawn from sight, the last alone of the increasing series being issued to be copied afresh. In this manner series were formed of successive copies, each of which was intended to be a faithful representation of the one immediately preceding it. Still, and it is to this that I wish to particularly draw attention although no two adjacent sketches exhibit very marked differences, the extremes of the series show hardly any resemblance to one another, and, if seen apart from their series, would certainly not be recognised as the same design, or as being in any way related to one another. The examples here given (see Plate I.) will serve to illustrate this, and, humorous and even frivolous though they appear, afford good examples of the unconscious variation of a design, the result of want of skill. The successive sketches are numbered from 1 to 14 in the order in which they were made. No. 1 is a sketch representing a snail crawling over a twig. In the course of six successive copyings the design had lost its meaning; by No. 10 the shell of the snail had left the body of the mollusc and had 'crawled' up the twig, the hinder end of the snail becoming intimately associated with the base of the twig. No. 12 is a copy made by a skilled artist, who was asked to 'interpret' the design at this stage, and to show in his sketch what he thought it was intended to re- present. The next copyist, not being able to make anything of the design when viewed the right way up, reversed it and proceeded with satisfaction to copy it upside down, under the impression that he was reproducing a 'bird' design; so also in No. 14, and in the succeeding copies which are here omitted, this interpretation was retained. This truly is 'evolution made easy'! The bird can here be traced back to its gastropodous molluscan prototype, through a continuous series of developmental changes of the simplest nature! The whole metamorphosis requires but fourteen of these changes, covering a period, say, of a day or so, and there is your bird-like form still irredeemably connected with its humble prototype, however much it may, if seen by itself, appear to scorn such an alliance.

It is interesting to trace what becomes of the various portions of the design individually. The large end of the twig becomes the bird's head, the growth-rings supplying the eye; the snail's body remains as that of the bird; the snail's head, with the prominent 'eye-stalks,' degenerates into the forked tail of the bird; the shell of the snail into an unwieldy and unnecessary wart upon the, shall we call them, 'trousers' which were once the branching end of the twig. One more example may serve to further illustrate the workings of the process {see Plate II.).

A sketch... of the head of the Patroclus, of the Aegina Marbles, was submitted to be copied. In the course of eight or nine successive copyings, the Greek warrior became metamorphosed into a female figure, a sort of helmeted Minerva, the helmet being much modified from the original type. But the most interesting point to notice is the ultimate fate of the strongly-emphasised lines representing the muscles and the collar-bones, a prominent representation of which is so characteristic of Greek art. The lines of the collar-bones droop lower and lower at the centre, till at length they form together an elongated V, giving to the later copyists the impression of a cloak thrown loosely over the shoulders, and even enveloping the arms, which were originally free and distinct, the muscle lines going to supply folds in the mantle. At No. 9, as at No. 12 in the last series, a skilled artist was asked to interpret the design.

These two examples will be sufficient to show the importance of this process in the production of variations upon a given design, and how an original idea may, through lack of skill, become modified and completely lost, while a new suggestion of its meaning may arise and be accepted, this again in later stages tending to be obscured and to become meaningless. The process of 'degradation' is hastened or retarded according to the greater or less skill displayed, but the design is unconsciously varied, and in no case has there been any intention to make the copy differ from the thing copied. Although, in order to illustrate this point with series whose continuity could not be doubted, I have not drawn my examples from actual savage art, still the workings of this process are often very apparent in the art of the lower races, as also in that of civilised peoples.

It would appear that this process was an important factor in producing the remarkable series of variations from the original type to be noticed in the ancient Gaulish and British imitations of foreign coins, such as the gold stater of Philip of Macedon, as has been so well described by Dr. John Evans.

Thus it may be stated that, in the early condition of man's culture, conventional or fanciful design, which is usually the result of such variation, has been to a great extent unconsciously evolved from realistic representation, and the passage from the one to the other has been by easy stages, by successive slight changes.

It is obvious that, if instead of actually copying a design, the reproduction is made from memory the process of unconscious variation will be greatly hastened.

(b) Conscious Variation—This unintentional variation of design is, however, frequently, usually in fact, accelerated by another process, which I have called conscious variation, that is to say, the desire to vary or improve upon the design copied. While the two processes may be associated, each contributing to the changes effected, conscious variation is frequently to all intents and purposes the sole agent. Some resemblance to the accepted type or model is retained, but there is no idea of slavishly adhering to the original in detail. This usually results in some particular portion or portions of the original design being specially emphasised, and made thus to develop at the expense of the remainder.

Conscious variation may act in many different ways, and be dictated by a variety of circumstances, and I now propose to give a number of illustrative examples of its importance in the development of ornamental designs, and the evolution of conventional designs from more or less realistic originals.

Balfour 'Evolution of Decorative Art' Fig. 10

On the shafts of many of the very elaborate spears or lances from the islands of the Solomon group, Melanesia—especially those from Bougainville Straits and St. Christoval—is seen a curious pattern, carved in low relief at the base of the long multibarbed head. This pattern varies to a considerable extent, as may be seen from the examples here given, but in examining a number together it becomes obvious that they are all modifications of one original design, that they are all traceable to one primary conception, that they are in fact survivals of various stages of variation. By a process of associating together those of them which most resemble each other, a series is formed which at one end is composed of designs such as that shown in Fig. l0a, while at the other end we find patterns of the kind shown in Fig. 10b. Between these extremes are various intermediate forms, which resemble the extremes more and more as they approach them, and which are more 'generalised' towards the centre.

From such an arrangement we learn that the original type was a representation of the human form, not 'divine' but grotesque, though still recognisable as such. This is represented in Fig. 10a, which shows the body with arms and legs fully represented, though the head is detached; this latter, it will be noticed, is chiefly remarkable for the very prominent angular mouth. Fig. 10b shows a modification of this conception; the body has been split up in order to increase the decorative effect, the arms and legs coalesce with those of the corresponding figure on the other side; the face too is modified. Fig. l0c shows another modification of the design, in which the various parts are still more conventionalised, the reduction of the body and limbs has proceeded further, the former a mere central patch, the latter represented by lines down the sides which meet each other across the base, forming thus a kind of frame to the lower part of the design. The mouth, absent in 10b, is here extremely prominent, being in fact reduplicated to form a bold double chevron. In Fig. 10d the 'mouth' is triple, and by far the most important part of the design, the central 'body' patch and rectangular frame still surviving. Fig. 10e represents a perfectly conventional pattern derived from the design; the face is suppressed, while the all-devouring mouth is quadrupled, and, for purposes of symmetry, a corresponding number of 'mouths' are added turned the other way, forming an elegant though meaningless chevron pattern. The rectangular remains of the body and legs have, curiously enough, survived the various changes which have led up to the fanciful pattern, and remain to emphasise the alliance of the last stages to the first.

Such a series as this cannot pretend to illustrate the continuous succession of variations which have led up to the conventionalised pattern, but it shows, in a general way, the manner in which this has been gradually arrived at. There are numerous other variations upon this design, which branch off at different stages. The actual genealogy of these various connected patterns derived from the original design of the human form upon these spears, would have to be represented in the form of a tree, with main stem, branches, and sub-branches. ...

The series which I have just given may be taken as illustrating the amplification of a design, and the reduction of realism for increased decorative effect, and I will now give an example to show how a design, originally intended to be both realistic and decorative, may tend to disappear little by little, and perhaps vanish altogether.

Balfour 'Evolution of Decorative Art' Fig. 13

A very large proportion of the pottery vessels made by the ancient inhabitants of Peru for holding water or other purposes, are shaped so as to represent animal and other forms more or less conventionalised, and of these a considerable number show the human form more or less grotesquely portrayed. By examination of a large series the stages by which the original quasi-realistic design has become modified, and has gradually vanished, can to a great extent be made out. The following examples, selected from the Peruvian pottery in the Pitt Rivers Collection, will serve to illustrate, in a broad and general way, the transitions. The series cannot be taken as representing the history and fate of the particular form of human representation shown in No. I., but as illustrating the modification of the human form conception as applied to this class of pottery as a whole, involving a variety of modes of treatment of this theme. I am unable to say to what extent these variations in style are local differences within the area of Peru, or whether each style was intended to convey a special symbolic meaning. ...

There can be little doubt that, in the series just given, the disappearance of the details of the original design has been effected to a great extent by un-skilfulness or carelessness on the part of the potters, who, while to a certain extent desirous of perpetuating a design which was generally recognised as suited to this class of vessel, successively allowed the various attributes to drop out, leaving them to the imagination, as it was easier to do this than to represent in full a somewhat complicated design, in a material which in any case presented difficulties to the manipulator.

...In further illustration of the process of conscious variation, one or two examples from modern civilised art may be given.

Where a particular design has a special meaning, which renders its frequent repetition desirable, we find it often introduced under circumstances which do not admit of realism, and there will be a combination of 'symbolic' with 'aesthetic' representation that is to say, there will be an endeavour to increase the ornamental effect, at the same time retaining sufficient of the attributes of the original design to enable this still to be recognised in its conventionalised form. In Chinese art the bat, considered to be the precursor of happiness and prosperity, forms a frequent theme for decorative design, and variations upon this animal pattern occur upon a great variety of objects. I give here some of the variations which show greater or less degrees of conventionalisation. ...

Occasionally there is a definite and valid reason for specially emphasising some particular portion of a design. That portion may have a special significance, which renders it the most important feature. In reproducing such a design again and again, the tendency would always be to preserve and make the most of this important detail, and the rest of the design would tend to sink into insignificance before its encroachments, perhaps to disappear altogether, leaving the principal feature master of the field.

When one small boy, if I may digress for a moment, puts out his tongue at another, it is understood that he desires to convey the impression of feelings of contempt and defiance towards the other. The action is a simple one, but conveys a considerable amount of information, being, moreover, an incentive to action on the other's part. It may be said there- fore that, for the time being—happily usually only a brief period—the protruded tongue, from the implied emotion, becomes the most important feature of the boy's face; and, even supposing that the eyes and nose were concealed, it would still convey its meaning, and serve to stimulate the latent energy of boy number two. This digression is not without its purpose, as this precise point finds expression in savage art. The ceremonial staves, hani, used by the Maori chiefs of New Zealand, are decorated at their upper extremities with carved grotesques of the human face, with the eyes made of discs of haliotis shell (Fig. 23a). The most prominent portion of the carved design is the enormous lanceolate tongue, which is shown protruded from the mouth, usually covered with an elaborate carved scroll-work. The raison d'etre of this greatly emphasised tongue lies in the fact that with the Maori warriors, as with our small boys, the protrusion of this organ indicates defiance and contempt, and this form of expression of these emotions is of considerable importance with this warlike people, the cultivation of this accomplishment forming an important part of the warrior's training. ...

The special retention and amplification of certain important portions of symbolic designs is very characteristic of mythological and totemic representations in savage art. The Haidas of British Columbia, for example, frequently decorate large spaces, the sides of wooden coffins, boxes, and other objects, by covering them completely with the eyes of the 'thunder-bird,' or hands, or the two intermingled, painted upon the surfaces with highly decorative effect, the design at the same time retaining its symbolic significance. ...

The history and fate of symbolic designs has recently received much of the attention which its study deserves, but a very wide field yet remains unexplored, inviting investigation with the promise of most interesting results, which will amply repay those who are willing to examine the material with sufficient care. Quite recently Dr. E. B. Tylor has been able to solve the mystery of the designs upon Assyrian and other ancient monuments, representing winged human figures holding in their hands cone-like objects, which they extend towards a much conventionalised palm-tree. A careful study of these designs goes far towards proving that they represent the artificial fertilisation of the palm-tree ... The symbolical nature of these representations of the ' tree of life led to their repetition again and again through successive ages, and their adoption by other races. Modifications in the details and the increasing conventionalisation of the design for ornamental uses tended to obscure its meaning, until finally this became lost, the unity was broken up, and the winged figures became separated from the palm-tree, though they still continued to survive as distinct designs. From the palm a variety of patterns was developed in different parts of the world, amongst others the so-called 'honeysuckle' or 'palmette' in Greek art; while to the winged figures may be traced, through a long genealogical succession, the winged genii of Greek, Etruscan, and Roman art, and the angels of Christianity. The two separated portions of the old design might meet again in some accidental association, but it would be without any sign of recognition. ...

It is impossible to estimate how many of the decorative designs with which we are surrounded, and which we are accustomed to regard as mere ornamental patterns, could be traced, were the material forthcoming, to originals having a definite significance, it may be one of sacred import, to the minds of those for whose benefit they were designed. Symbols of the various religious beliefs and institutions of Man in his different stages of culture have been repeated again and again, ever with an increasing tendency to vary from the original, have been transmitted from one people to another, till all semblance to the original and all knowledge of its significance have vanished. The meaningless and fanciful pattern which is the net result of all this continues, partly from the force of habit, to be applied to aesthetic purposes, and may frequently become, as decoration, associated with an object to which the original design would have been singularly inappropriate.

In spite of this ever present tendency to vary accepted designs, and the many causes which produce changes, one of the prominent characteristics of savage art is the persistent manner in which certain types of designs or themes are adhered to; the conservative nature of the mind of the savage is well exemplified in his art of decoration. ...

Among the very many external influences which act upon designs and upset their stability, one of the most constant is that of one design over another. This may be in many cases merely the application of the more decorative characteristics of one design to give increased effect to another, which is perhaps itself lacking in this respect. Frequently, however, this interchange of characteristics would appear less purposeful, as it is difficult to see that anything is gained thereby. ...

It frequently happens when a design is repeated, so that two or more are brought into juxtaposition and symmetrically disposed with regard to one another, that an entirely new idea is produced, suggesting a new 'theme' for use as ornament, which may then continue to develop and be modified upon lines different from those suggested by the design before it became thus reduplicated. ...

In the decoration of objects useful in the every-day life of savages—such as weapons, tools, and the like—as the ornamentation has, in its earlier stages at least, been often necessarily influenced, if not directly suggested, by the form or function of the object, it is usually admirably adapted to the latter. We thus find a true balance of ornament and useful form, very pleasing to the eye, and suggesting considerable artistic taste on the part of the lowly cultured artists. Not only do we find this to be very generally the case with the decorated implements, but it also holds good in the case of the beautiful design is often tattooed upon different portions of the body...

Before proceeding to discuss the natural history of decorative art from another point of view, it may be well to recall, by means of a brief synoptical table, the main points connected with the evolution of this art, as illustrated in the preceding pages. The chief stages, following the first appreciation of peculiarities or ornamental effects, either natural or produced as accidents in manufacture, may be summed up as follows :

I. Adaptive:—The appreciation of curious or decorative effects occurring in nature or as accidents in manufacture, and the slight increasing of the same by artificial means in order to augment their peculiar character or enhance their value as ornament.

II. Creative:—The artificial production oi similar effects where these do not occur; imitation or copying.

III. Variative:—Gradual metamorphosis of designs by-

(a) Unconscious Variation in which the changes are not intentional, but are due to want of skill or careless copying, difficulty of material, or reproducing from memory.

{b) Conscious Variation in which the changes are intentional, and may be made to serve some useful purpose (e.g. marks of ownership), or to increase an ornamental effect; to emphasise some specially important feature in a symbolic design; to adapt the same design to a variety of objects or spaces; by the development of a new idea from the modification of a pre-existing design; etc.

The effects of 'successive copying' are paramount in creating variations upon established designs whether unconsciously or consciously.

Conscious and unconscious variation frequently, usually in fact, act together, the changes being hastened by an association of unskilful copying and a desire to vary.

[Balfour then goes on to consider Sculpture and Delineation and Graphic arts and also natural forms which are interpreted with meaning, like the root of a mandrake resembling a human figure].

After 128 pages Balfour ends with a appropriately visual joke.


Pitt-Rivers owned a copy of this book which appears on page 22 of the catalogue of his library [CUL Add.1944/10].

Bibliography for this article

Balfour, Henry. 1893. 'The Evolution of Decorative Art: An essay upon its origin and development as illustrated by the art of modern races of mankind' London: Rivington, Percival and Co.

Balfour, Henry, 1894. ‘Evolution in Decorative Art’ in the Journal of the Society of Arts, April 1894

Pitt-Rivers, A.H.L.F. 1884. 'Address delivered at the Annual Meeting of the Dorchester School of Art, February 1884' Dorset Chronicle, February 7 1884 (pagination unknown)

[Transcribed by AP as part of the Rethinking Pitt-Rivers project, June 2010]

Balfour 'Evolution of Decorative Art' page 128

prm logo