the masses that go to make up a picture have variety in their shape, their tone values, their edges, in texture or quality, and in gradation. quite a formidable list, but each of these particulars has some rhythmic quality of its own about which it will be necessary to say a word.

variety of shape.

as to variety of shape, many things that were said about lines apply equally to the spaces enclosed by them. it is impossible to write of the rhythmic possibilities that the infinite variety of shapes possessed by natural objects contain, except to point out how necessary the study of nature is for this. variety of shape is one of the most difficult things to invent, and one of the commonest things in nature. however imaginative your conception, and no matter how far you may carry your design, working from imagination, there will come a time when studies from nature will be necessary if your work is to have the variety that will give life and interest. try and draw from imagination a row of elm trees of about the same height and distance apart, and get the variety of nature into them; and you will see how difficult it is to invent. on examining your work you will probably discover two or three pet forms repeated, or there may be only one. or try and draw some cumulus clouds from imagination, several groups of them 186across a sky, and you will find how often again you have repeated unconsciously the same forms. how tired one gets of the pet cloud or tree of a painter who does not often consult nature in his pictures. nature is the great storehouse of variety; even a piece of coal will suggest more interesting rock-forms than you can invent. and it is fascinating to watch the infinite variety of graceful forms assumed by the curling smoke from a cigarette, full of suggestions for beautiful line arrangements. if this variety of form in your work is allowed to become excessive it will overpower the unity of your conception. it is in the larger unity of your composition that the imaginative faculty will be wanted, and variety in your forms should always be subordinated to this idea.

nature does not so readily suggest a scheme of unity, for the simple reason that the first condition of your picture, the four bounding lines, does not exist in nature. you may get infinite suggestions for arrangements, and should always be on the look out for them, but your imagination will have to relate them to the rigorous conditions of your four bounding lines, and nature does not help you much here. but when variety in the forms is wanted, she is pre-eminent, and it is never advisable to waste inventive power where it is so unnecessary.

but although nature does not readily suggest a design fitting the conditions of a panel her tendency is always towards unity of arrangement. if you take a bunch of flowers or leaves and haphazard stuff them into a vase of water, you will probably get a very chaotic arrangement. but if you leave it for some time and let nature have 187a chance you will find that the leaves and flowers have arranged themselves much more harmoniously. and if you cut down one of a group of trees, what a harsh discordant gap is usually left; but in time nature will, by throwing a bough here and filling up a gap there, as far as possible rectify matters and bring all into unity again. i am prepared to be told this has nothing to do with beauty but is only the result of nature's attempts to seek for light and air. but whatever be the physical cause, the fact is the same, that nature's laws tend to pictorial unity of arrangement.

variety of tone values

it will be as well to try and explain what is meant by tone values. all the masses or tones (for the terms are often used interchangeably) that go to the making of a visual impression can be considered in relation to an imagined scale from white, to represent the lightest, to black, to represent the darkest tones. this scale of values does not refer to light and shade only, but light and shade, colour, and the whole visual impression are considered as one mosaic of masses of different degrees of darkness or lightness. a dark object in strong light may be lighter than a white object in shadow, or the reverse: it will depend on the amount of reflected light. colour only matters in so far as it affects the position of the mass in this imagined scale of black and white. the correct observation of these tone values is a most important matter, and one of no little difficulty.

the word tone is used in two senses, in the first place when referring to the individual masses as to their relations in the scale of "tone values"; and secondly when referring to the musical relationship of these values to a oneness of tone idea 188governing the whole impression. in very much the same way you might refer to a single note in music as a tone, and also to the tone of the whole orchestra. the word values always refers to the relationship of the individual masses or tones in our imagined scale from black to white. we say a picture is out of value or out of tone when some of the values are darker or lighter than our sense of harmony feels they should be, in the same way as we should say an instrument in an orchestra was out of tone or tune when it was higher or lower than our sense of harmony allowed. tone is so intimately associated with the colour of a picture that it is a little difficult to treat of it apart, and it is often used in a sense to include colour in speaking of the general tone. we say it has a warm tone or a cold tone.

there is a particular rhythmic beauty about a well-ordered arrangement of tone values that is a very important part of pictorial design. this music of tone has been present in art in a rudimentary way since the earliest time, but has recently received a much greater amount of attention, and much new light on the subject has been given by the impressionist movement and the study of the art of china and japan, which is nearly always very beautiful in this respect.

this quality of tone music is most dominant when the masses are large and simple, when the contemplation of them is not disturbed by much variety, and they have little variation of texture and gradation. a slight mist will often improve the tone of a landscape for this reason. it simplifies the tones, masses them together, obliterating many smaller varieties. i have even heard of the tone 189of a picture being improved by such a mist scrambled or glazed over it.

plate xliii.

monte solaro capri

study on brown paper in charcoal and white chalk.

the powder on a lady's face, when not over-done, is an improvement for the same reason. it simplifies the tones by destroying the distressing shining lights that were cutting up the masses; and it also destroys a large amount of half tone, broadening the lights almost up to the commencement of the shadows.

tone relationships are most sympathetic when the middle values of your scale only are used, that is to say, when the lights are low in tone and the darks high.

they are most dramatic and intense when the contrasts are great and the jumps from dark to light sudden.

the sympathetic charm of half-light effects is due largely to the tones being of this middle range only; whereas the striking dramatic effect of a storm clearing, in which you may get a landscape brilliantly lit by the sudden appearance of the sun, seen against the dark clouds of the retreating storm, owes much of its dramatic quality to contrast. the strong contrasts of tone values coupled with the strong colour contrast between the warm sunlit land and the cold angry blue of the storm, gives such a scene much dramatic effect and power.

the subject of values will be further treated in dealing with unity of tone.

variety in quality and texture

variety in quality and nature is almost too subtle to write about with any prospect of being understood. the play of different qualities and textures in the masses that go to form a picture must be appreciated at first hand, and little can be written about it. oil paint is capable of almost unlimited variety in this 190way. but it is better to leave the study of such qualities until you have mastered the medium in its more simple aspects.

the particular tone music of which we were speaking is not helped by any great use of this variety. a oneness of quality throughout the work is best suited to exhibit it. masters of tone, like whistler, preserve this oneness of quality very carefully in their work, relying chiefly on the grain of a rough canvas to give the necessary variety and prevent a deadness in the quality of the tones.

but when more force and brilliancy are wanted, some use of your paint in a crumbling, broken manner is necessary, as it catches more light, thus increasing the force of the impression. claude monet and his followers in their search for brilliancy used this quality throughout many of their paintings, with new and striking results. but it is at the sacrifice of many beautiful qualities of form, as this roughness of surface does not lend itself readily to any finesse of modelling. in the case of claude monet's work, however, this does not matter, as form with all its subtleties is not a thing he made any attempt at exploiting. nature is sufficiently vast for beautiful work to be done in separate departments of vision, although one cannot place such work on the same plane with successful pictures of wider scope. and the particular visual beauty of sparkling light and atmosphere, of which he was one of the first to make a separate study, could hardly exist in a work that aimed also at the significance of beautiful form, the appeal of form, as was explained in an earlier chapter, not being entirely due to a visual but to a mental perception, into which the sense 191of touch enters by association. the scintillation and glitter of light destroys this touch idea, which is better preserved in quieter lightings.

there is another point in connection with the use of thick paint, that i don't think is sufficiently well known, and that is, its greater readiness to be discoloured by the oil in its composition coming to the surface. fifteen years ago i did what it would be advisable for every student to do as soon as possible, namely, make a chart of the colours he is likely to use. get a good white canvas, and set upon it in columns the different colours, very much as you would do on your palette, writing the names in ink beside them. then take a palette-knife, an ivory one by preference, and drag it from the individual masses of paint so as to get a gradation of different thicknesses, from the thinnest possible layer where your knife ends to the thick mass where it was squeezed out of the tube. it is also advisable to have previously ruled some pencil lines with a hard point down the canvas in such a manner that the strips of paint will cross the lines. this chart will be of the greatest value to you in noting the effect of time on paint. to make it more complete, the colours of several makers should be put down, and at any rate the whites of several different makes should be on it. as white enters so largely into your painting it is highly necessary to use one that does not change.

the two things that i have noticed are that the thin ends of the strips of white have invariably kept whiter than the thick end, and that all the paints have become a little more transparent with time. the pencil lines here come in useful, as they can be seen through the thinner portion, and show to what 192extent this transparency has occurred. but the point i wish to emphasise is that at the thick end the larger body of oil in the paint, which always comes to the surface as it dries, has darkened and yellowed the surface greatly; while the small amount of oil at the thin end has not darkened it to any extent.

claude monet evidently knew this, and got over the difficulty by painting on an absorbent canvas, which sucks the surplus oil out from below and thus prevents its coming to the surface and discolouring the work in time. when this thick manner of painting is adopted, an absorbent canvas should always be used. it also has the advantage of giving a dull dry surface of more brilliancy than a shiny one.

although not so much as with painting, varieties of texture enter into drawings done with any of the mediums that lend themselves to mass drawing; charcoal, conté crayon, lithographic chalk, and even red chalk and lead pencil are capable of giving a variety of textures, governed largely by the surface of the paper used. but this is more the province of painting than of drawing proper, and charcoal, which is more painting than drawing, is the only medium in which it can be used with much effect.

variety of edges.

there is a very beautiful rhythmic quality in the play from softness to sharpness on the edges of masses. a monotonous sharpness of edge is hard, stern, and unsympathetic. this is a useful quality at times, particularly in decorative work, where the more intimate sympathetic qualities are not so much wanted, and where the harder forms go better with the architectural surroundings of which your painted decoration should form a part. on the other hand, a monotonous softness of edge is very weak and feeble-looking, and 193too entirely lacking in power to be desirable. if you find any successful work done with this quality of edge unrelieved by any sharpnesses, it will depend on colour, and not form, for any qualities it may possess.

some amount of softness makes for charm, and is extremely popular: "i do like that because it's so nice and soft" is a regular show-day remark in the studio, and is always meant as a great compliment, but is seldom taken as such by the suffering painter. but a balance of these two qualities playing about your contours produces the most delightful results, and the artist is always on the look out for such variations. he seldom lets a sharpness of edge run far without losing it occasionally. it may be necessary for the hang of the composition that some leading edges should be much insisted on. but even here a monotonous sharpness is too dead a thing, and although a firmness of run will be allowed to be felt, subtle variations will be introduced to prevent deadness. the venetians from giorgione's time were great masters of this music of edges. the structure of lines surrounding the masses on which their compositions are built were fused in the most mysterious and delightful way. but although melting into the surrounding mass, they are always firm and never soft and feeble. study the edge in such a good example of the venetian manner as the "bacchus and ariadne" at the national gallery, and note where they are hard and where lost.

there is one rather remarkable fact to be observed in this picture and many venetian works, and this is that the most accented edges are reserved for unessential parts, like the piece of white drapery 194on the lower arm of the girl with the cymbals, and the little white flower on the boy's head in front. the edges on the flesh are everywhere fused and soft, the draperies being much sharper. you may notice the same thing in many pictures of the later venetian schools. the greatest accents on the edges are rarely in the head, except it may be occasionally in the eyes. but they love to get some strongly-accented feature, such as a crisply-painted shirt coming against the soft modelling of the neck, to balance the fused edges in the flesh. in the head of philip iv in our national gallery the only place where velazquez has allowed himself anything like a sharp edge is in the high lights on the chain hanging round the neck. the softer edges of the principal features in these compositions lend a largeness and mystery to these parts, and to restore the balance, sharpnesses are introduced in non-essential accessories.

in the figure with the white tunic from velazquez's "surrender of breda," here reproduced, note the wonderful variety on the edges of the white masses of the coat and the horse's nose, and also that the sharpest accents are reserved for such non-essentials as the bows on the tunic and the loose hair on the horse's forehead. velazquez's edges are wonderful, and cannot be too carefully studied. he worked largely in flat tones or planes; but this richness and variety of his edges keeps his work from looking flat and dull, like that of some of his followers. i am sorry to say this variety does not come out so well in the reproduction on page 194 [transcribers note: plate xliv] as i could have wished, the half-tone process having a tendency to sharpen edges rather monotonously.

this quality is everywhere to be found in 195nature. if you regard any scene pictorially, looking at it as a whole and not letting your eye focus on individual objects wandering from one to another while being but dimly conscious of the whole, but regarding it as a beautiful ensemble; you will find that the boundaries of the masses are not hard continuous edges but play continually along their course, here melting imperceptibly into the surrounding mass, and there accentuated more sharply. even a long continuous line, like the horizon at sea, has some amount of this play, which you should always be on the look out for. but when the parts only of nature are regarded and each is separately focussed, hard edges will be found to exist almost everywhere, unless there is a positive mist enveloping the objects. and this is the usual way of looking at things. but a picture that is a catalogue of many little parts separately focussed will not hang together as one visual impression.

plate xliv.

part of the surrender of breda. by velazquez

note the varied quantity of the edge in white mass of tunic. (the reproduction does not unfortunately show this as well as the original.)

photo anderson

in naturalistic work the necessity for painting to one focal impression is as great as the necessity of painting in true perspective. what perspective has done for drawing, the impressionist system of painting to one all-embracing focus has done for tone. before perspective was introduced, each individual object in a picture was drawn with a separate centre of vision fixed on each object in turn. what perspective did was to insist that all objects in a picture should be drawn in relation to one fixed centre of vision. and whereas formerly each object was painted to a hard focus, whether it was in the foreground or the distance, impressionism teaches that you cannot have the focus in a picture at the same time on the foreground and the distance.

196of course there are many manners of painting with more primitive conventions in which the consideration of focus does not enter. but in all painting that aims at reproducing the impressions directly produced in us by natural appearances, this question of focus and its influence on the quality of your edges is of great importance.

something should be said about the serrated edges of masses, like those of trees seen against the sky. these are very difficult to treat, and almost every landscape painter has a different formula. the hard, fussy, cut-out, photographic appearance of trees misses all their beauty and sublimity.

there are three principal types of treatment that may serve as examples. in the first place there are the trees of the early italian painters, three examples of which are illustrated on page 197 [transcribers note: diagram xxiii]. a thin tree is always selected, and a rhythmic pattern of leaves against the sky painted. this treatment of a dark pattern on a light ground is very useful as a contrast to the softer tones of flesh. but the treatment is more often applied nowadays to a spray of foliage in the foreground, the pattern of which gives a very rich effect. the poplar trees in millais' "vale of rest" are painted in much the same manner as that employed by the italians, and are exceptional among modern tree paintings, the trees being treated as a pattern of leaves against the sky. millais has also got a raised quality of paint in his darks very similar to that of bellini and many early painters.

giorgione added another tree to landscape art: the rich, full, solidly-massed forms that occur in his "concert champêtre" of the louvre, reproduced on page 151 [transcribers note: plate xxxiii]. in this picture you may see both types 197of treatment. there are the patterns of leaves variety on the left and the solidly-massed treatment on the right.

diagram xxiii.

examples of early italian treatment of trees

a. from pictures in oratorio di s. ansano. "il trionfo dell' amore," attributed to botticelli.

b. from "l'annunziazione," by botticelli, uffizi, florence.

c. from "la vergine," by giovanni bellini in the accademia, venice.

corot in his later work developed a treatment that has been largely followed since. looking at trees with a very wide focus, he ignored individual leaves, and resolved them into masses of tone, 198here lost and here found more sharply against the sky. the subordinate masses of foliage within these main boundaries are treated in the same way, resolved into masses of infinitely varying edges. this play, this lost-and-foundness at his edges is one of the great distinguishing charms of corot's trees. when they have been painted from this mass point of view, a suggestion of a few leaves here and a bough there may be indicated, coming sharply against the sky, but you will find this basis of tone music, this crescendo and diminuendo throughout all his later work (see illustration, page 215 [transcribers note: diagram xxvi]).

these are three of the more extreme types of trees to be met with in art, but the variations on these types are very numerous. whatever treatment you adopt, the tree must be considered as a whole, and some rhythmic form related to this large impression selected. and this applies to all forms with serrated edges: some large order must be found to which the fussiness of the edges must conform.

the subject of edges generally is a very important one, and one much more worried over by a master than by the average student. it is interesting to note how all the great painters have begun with a hard manner, with edges of little variety, from which they have gradually developed a looser manner, learning to master the difficulties of design that hard contours insist on your facing, and only when this is thoroughly mastered letting themselves develop freely this play on the edges, this looser handling.

for under the freest painting, if it be good, there will be found a bed-rock structure of well-constructed 199masses and lines. they may never be insisted on, but their steadying influence will always be felt. so err in your student work on the side of hardness rather than looseness, if you would discipline yourself to design your work well. occasionally only let yourself go at a looser handling.

variety of gradiation.

variety of gradation will naturally be governed largely by the form and light and shade of the objects in your composition. but while studying the gradations of tone that express form and give the modelling, you should never neglect to keep the mind fixed upon the relation the part you are painting bears to the whole picture. and nothing should be done that is out of harmony with this large conception. it is one of the most difficult things to decide the amount of variety and emphasis allowable for the smaller parts of a picture, so as to bring all in harmony with that oneness of impression that should dominate the whole; how much of your scale of values it is permissible to use for the modelling of each individual part. in the best work the greatest economy is exercised in this respect, so that as much power may be kept in reserve as possible. you have only the one scale from black to white to work with, only one octave within the limits of which to compose your tone symphonies. there are no higher and lower octaves as in music to extend your effect. so be very sparing with your tone values when modelling the different parts.