AN 
 
 ESSAY 
 
 ON A 
 
 NEW THEORY OF COLOURS, 
 
 AND ON 
 
 COMPOSITION IN GENERAL,? 
 
 ILLUSTRATED BY 
 
 COLOURED BLOTS 
 
 SHEWING THE 
 
 APPLICATION OF THE THEORY 
 
 TO COMPOSITION OF 
 
 FLOWERS, LANDSCAPES, FIGURES, &c. 
 
 IN THREE PARTS, 
 
 BY M. OARTSIDE* 
 
 '• ~ >. 
 
 THE SECOND EDITION. 
 
 “RETURN FAIR COLOURING!- 
 
 -CHASTE SEDUCER! SAY, 
 
 WIIAT LAWS CONTROUL THEE, AN t) WEAT POWERS OBEY. 
 
 DU FRESNOY. 
 
 LonOan: 
 
 Printed by J. Barfield, Wardour-Street, 
 
 FOR 
 
 T. GARDINER, P1UNC ES-STREET, CAVENDISII-SQ0ARE, 
 
 W. MILLER, ALBEMARLE-STREET, AND I. AND A. ARCII, CORNIJILL. 
 
 1808. 
 
 V 
 
 I 
 
 I 
 
 
TO THE RIGHT HONOURABLE 
 
 LADY SOPHIA GREY. 
 
 T? H E permission to address this little work to 
 Your Ladyship, is not the least of the many favours you have 
 conferred upon me. 
 
 Duty to my Pupils, and some hope that it may be of 
 use to accelerate their progress in an Art you encourage, induced me 
 to undertake it; and it is both my pleasure and my pride to have 
 this public opportunity of assuring Your Ladyship that I am, with 
 the greatest gratitude and respect. 
 
 Your most obliged 
 
 and most devoted Servant, 
 
 M. GARTSIDE. 
 
PREFACE 
 
 TO THE 
 
 SECOJVE EJBITIOJV' 
 
 As a reference to the former Edition ot this Work is fre¬ 
 quently made in a Work entitled Ornamental Groups, lately pub¬ 
 lished; and as the attention of the possessors of that Work 
 may be directed to this Essay, the Author thinks it would be 
 treating the Public with disrespect, to refer them to a Work 
 that might be in any respect rendered more perfect: She has, 
 therefore, studiously attended to the different ciitisisms passed 
 upon the former Edition, both by indifferent people, and also 
 by her particular Friends. The principle upon which the theory 
 is founded has been approved of; but previous to its being now a 
 second time presented to public notice, it has undergone a 
 careful revisal as to style ; some parts that were thought obscure 
 
6 
 
 she has endeavoured to express in a clearer manner; and others 
 that did not discuss the subject sufficiently, have been enlarged. 
 The Work is also now presented to the Public, it is hoped, in 
 a more acceptable manner as to Price; which in the first edition, 
 owing to the Author’s inexperience in the management of public 
 Works, was fixed too high; although the expence attending it 
 fully justified the Publisher’s estimate. In order, however, to re¬ 
 medy the error, as far as it is possible, the addition now published 
 may be had of the publisher T. Gardiner, for Half-a-Crown; 
 or it may be added to a second part of this Work, which will be 
 speedily published ; and in which the principles of the theory 
 will be extended to Landscapes, &c. 
 
 It may possibly appear presumptuous in one, who is pro¬ 
 fessedly only a flower painter, to offer any thing like instruction 
 to them who are practitioners in the higher branches of the Art; 
 but that is by no means the Author’s intention ; she means only 
 to present the principle to their notice, and to point out its ap¬ 
 plication to different objects, as well as' flowers, by blots adapted 
 to the purpose. 
 
 The difference of opinion alluded to in the following pages. 
 
7 
 
 respecting the number of prismatic colours, also whether each is 
 an original colour of itself, or only produced by a mixture of 
 the two contiguous ones, not being of importance to the subject 
 in question, I shall only observe, that painters reckon but three 
 primitive colours, Yellow, Red, and .Blue; and that an examina¬ 
 tion of the prismatic spectrum has led to the consideration and 
 knowledge of the different powers and qualities of these three 
 colours, which essentially differ from each other. 
 
 They may be considered as three distinct heads of three 
 distinct families of colours; each comprehending all the Yellows, 
 all the Reds, and all the Blues, from the lightest to the darkest tints; 
 and these families are no ways related to each other, but per¬ 
 form different offices in the prismatic system. Yellow illuminates, 
 Red warms, Blue is opposite to both these, and is of a cold re¬ 
 tiring nature; these at least are the effects that our corresponding 
 material colours produce when applied to the purposes of paint- 
 ting; and what makes it remarkable is, that although any one 
 of these three colours will unite and mix with either of the other 
 two separately, so as to form another pure brilliant prismatic 
 tint, yet no one of them will unite with the other two together. 
 
 B 
 
For when three are combined, the colours are debased; and 
 when mixed in a certain proportion, destroyed;* and black, or 
 darkness, the privation of light or colour produced. Without 
 entering at all into the question before alluded to, respecting the 
 prismatic colours, it is certain from experience, that even with 
 our gross materials, we can produce, by different mixtures of the 
 three prismatic colours, Yellow, Red, and Blue, colours similar to 
 all the other etherial prismatic tints, although in a degree very 
 far inferior in point of beauty. We see them arrayed by nature 
 in harmonious succession; and if we interrupt that succession, we 
 perceive the harmony is immediately destroyed; for the three 
 primitive colours being, as before observed, of different natures, 
 they will not associate together; but require the mediation of 
 a third colour, formed by a mixture of any two of these three 
 colours. It is that mixture, then, tvhich produces the harmony so 
 pleasing to the eye, and which conducts it insensibly from one 
 to another; we will say from Yellow io Green, and from Green to 
 Blue ; or in like manner, from Yellow to Orange, from Orange 
 to Red, from Red to Violet, and from Violet to Blue. Thus the 
 eye travels gradually from one to another of the three primitive 
 colours, without being offended by too quick a transition from one to 
 
 * For that proportion which produces black See page 12. 
 
9 
 
 the other, which would be the case, were it not for the intermediate 
 tints, Green, Orange, and Violet. 
 
 This explains the nature of the harmonizing tints; and I shall 
 endeavour to explain the contrasting tints, as both are made 
 the subjects of the following blots, and though particularly 
 specified in each, yet they may perhaps be better understood by 
 this explanation. 
 
 By the contrasting tint, is meant one which is the most 
 opposite in its nature from that to which you wish to oppose it: thus, 
 for instance, Yellow, Blue, and Red, as has already been shown, 
 are essentially different, from each other, but not so much so as 
 for any one of them to form the greatest contrast to the other; 
 a still greater may be produced to Yellow: for instance, by the 
 two together, Blue and Red, which form Purple or Violet. In this 
 colour, the opposition is double ; and when there is an equal quantity 
 of both these colours in the mixture, the strongest contrast to Yellow 
 is produced. Thus the intermediate tints, Violet, Green, and Orange, 
 form the strongest contrast to \ ellow, Red, and Blue. This may 
 be rendered more clear, by inspecting the following diagram, 
 where the three primitive colours being placed at the three 
 
10 
 
 angles of an equilateral triangle, present both the harmonizing 
 and contrasting tints to each colour at one view. If this diagram 
 be well considered, it will be found a certain guide for arrang¬ 
 ing colours in whatever subject they may be employed upon; 
 for nature is uniform in her laws; the objects may be changed 
 on which colours are thrown, but the distribution of light and 
 colour is the same throughout all her works. 
 
 Y 
 
 G / \ O 
 
 B ^-R 
 
 V 
 
 In this diagram, let Y, R, and B, represent the three primi¬ 
 tive colours, Yellow, Red, and Blue; if Y be produced to R, and 
 R to Y, the compound formed by the conjunction of these two 
 is Orange:—if B be produced to R, and R to B, it is Violet; 
 and by the conjunction of Y and B, the colour formed is Green. 
 The angles, therefore, represent the true primitive colours; the 
 central intermediate points the pure compounds. Orange, Green, 
 and Violet; and every spot between these show the different 
 shades of these compounds, each partaking most of the nature 
 of the primitive to which it approximates. 
 
tfe. . Y» V.i 
 
 ON 
 
 COLOURS, 
 
 THEIR ARRANGEMENT IN GROUPS. 
 
 It is not intended in the following pages to explain the nature, or cause 
 of colours, or to speak of the theory any further than as it relates to the 
 effect of colours in painting. To that end it is necessary to observe, that 
 the rays of the sun convey colour as well as light and heat; that there are 
 said to be seven original, or primary colours, which are all visible in the 
 rainbow, or through a prism; red, orange, yellow, green, blue, indigo, 
 violet. Notwithstanding this, some are of opinion, that there are in reality 
 only three original colours, red, yellow, and blue ; others think it probable 
 that there are four, including indigo, differing in opinion from philosophical 
 experiments on the rays of light. Be this as it may, it is certain, that by 
 compounding the first three colours the other four are produced, at least 
 they are very nearly imitated : for instance, if you mix red and yellow, they 
 produce orange; if yellow and blue, you get a green ; with red and blue, 
 a violet; and with blue and violet, indigo ; thus are four of the colours 
 produced by the other three. 
 
 I 
 
 C 2 
 
 r-. 
 
12 
 
 Red, yellow, and blue, when mixed together, will also produce brown, 
 and even black, when in the following proportions, “ fifteen parts blue, 
 
 “ five of red, and three of yellow;” or, without observing this nicety in 
 proportioning the parts, which is rather speculatively than practically just, 
 let your eye be your guide: it may be easily done. But Avhat may be 
 surprising to those unacquainted with the nature of colours, all the seven 
 mixed together will produce white; and though not a pure colour, yet one 
 that can be called nothing else; and if not useful as a colour, it is of im¬ 
 portance to know, that all the variety of tints, when mixed together, destroys 
 the strength of each other, and produces a weak unmeaning one. There 
 are other combinations also that produce white, as will be mentioned 
 hereafter. But from what little has been said, a person, unacquainted with 
 the theory, will have a more enlarged idea of the property of each colour, 
 and will be better able to tell what effect the mixing such and such colours, 
 on the pallet, will have. They will be aware, that the orange, though 
 formed to the hand by the colourman, is a mixture of the primitive colours, 
 red and yellow ; that green is a mixture of yellow and blue, and so on. It 
 will be obvious also, from very little practice, that by dividing, or separating 
 the particles of each colour, either by spreading them thinly with water upon 
 paper, or by mixing white with them, any tint may be produced, from the 
 fullest the colour itself will make, to the palest degree of it next to white; 
 and it must be observed, that in objects which are all of one colour, for 
 instance, all red or all blue, those parts of them that the light strikes full 
 upon, appear paler and lighter than the other parts, which, without being 
 in shade, are not so strongly enlightened. Now it is the business of a 
 painter to imitate this effect of light upon an object, either by the artifice of 
 spreading the colour thinly upon those parts, and leaving it thicker upon 
 the others, or by using the colour, mixed with white, to make it lighter, 
 which produces the same effect. How every colour and tint, to be met with 
 in nature, is to be formed, either the experience gained by practice, or the 
 seeing another person do it, must inform a learner; so many colours being 
 produced by working one tint over another on the paper. Therefore 
 we must leave this part of the business, and proceed to the arrangement 
 of colours. 
 
13 
 
 Colours may be divided into three classes, light, warm, and cold ; those 
 that incline to red and yellow, are termed warm, those that incline to blue 
 or black, cold; from the first two resembling in colour the sun and fire, 
 the latter night and darkness. And as the primitive colour of light is 
 white, all colours approaching to that, of course, are termed light. Now, 
 if two men, the one dressed in white, or a warm colour, yellow or red; the 
 other, in a cold one, blue or black ; stand at a distance, you will perceive 
 the man in white, &c. &c. much plainer than the man in blue, &c. inso¬ 
 much, that though both are equally distant, he that is in white, &c. will 
 even appear nearer to you : a plain proof that light colours come forward to 
 the eye, and cold ones retire from it. 
 
 It follows, therefore, if you wish to give roundness, or projection to any 
 part of an object, you would, from the foregoing observation, place a light, 
 and warm colour on that part you want to come forward, and a cold one on 
 that you wish to go back, and retire from the eye. There are other essential 
 points to consider in forming a composition of colours; one is, that as 
 there requires a certain degree of shade to set off the light, so there should 
 be a certain degree of cold colour to balance the warm tints, the one 
 answering to the other, so as to unite the two extremes gradually, and im¬ 
 perceptibly with one another; but still in such a manner as to have the 
 general tone a warm one : for nothing is more unpleasing than harsh cold 
 shades, nor any thing more inconsistent with a strong light. It is as if you 
 had placed the shades of evening with the light of noon, and there is an 
 entire difference between them; the shades of morning being warm, mellow, 
 and agreeable, whereas those of evening are cold, and disagreeable: and the 
 impropriety of introducing shades of that tint, with a warm light, must be 
 obvious; and yet, it may be easily done by those who do not consider the 
 cause of these shades, which, in the one case, is owing to the retiring light; 
 in the other, only to the interception of some object between them, and 
 the sun, which strongly enlightening the object, it casts a strong shade, 
 whose extremities are always tinctured with the colour of the object that 
 casts the shadow, and which being blended with the colour of that it falls 
 upon, produces that soft, mellow shade before spoken of. The first step 
 
14 
 
 towards a knowledge of composition in painting, is to understand perfectly 
 the strength,and management of each colour separately, which are their 
 proper contrasting, and which their harmonising tints; without this you 
 cannot dispose of them with judgment in a full group; for colours are to 
 the eye what the notes in music are to the ear, and will either produce 
 harmony or discord, according as they are managed. I shall endeavour to 
 explain what has, and will be said on the subject, by examples of different 
 groups of flowers, &c. ; but before I speak of particular groups, I must men¬ 
 tion the harmonising tints, without which it is impossible to form a group. 
 By harmonising tints, I mean those which should occupy the space between 
 the two extremes of light and shade, and of warm and cold, in short, those 
 that come the nearest to one another; as, for instance, yellow is the har¬ 
 monising tint to white*, as being nearest in lightness of tint; orange to 
 yellow, scarlet to orange, red to scarlet, blue to green, indigo to blue, violet 
 to blue, and black to indigo: and when a learner is become an adept in the 
 management of these colours in simple groups, their arrangement in a full 
 one will be no difficulty. I shall beg leave to style this part of the work 
 simple composition, and will now speak of each group separately. 
 
 * There is no doubt but that a weak degree of any other colour will harmonise with white, but 
 yellow is here spoken of in its primitive state. 
 
15 
 
 A COMPOSITION OF WHITE. 
 
 The true primitive colour of light, unmixed with any other substance, is 
 white. I shall therefore speak of this colour first. Its contrast,or opposite,is 
 of course black, or darkness. But as there never is a case in composition, 
 where a white object is left unqualified by shade, neither is there any,where 
 much pure black is required as a balance to it, so that a dark grey is almost a 
 balance sufficient: for though black must be admitted, it should be with great 
 care, and only in proportion to the pure touches of white in the object. But 
 though a very pleasing effect may, in some instances, be produced in compo¬ 
 sition by a warm tint, and its contrast, by blending the two extremes to¬ 
 gether, yet in painting from nature you are unavoidably obliged to employ 
 more; for instance, a white flower and its green leaves: and it may be im¬ 
 possible, with a strict attention to nature, if the composition is a slight one, 
 to give its balance,or contrast,any way but in the ground it is painted on; 
 and white objects should always be painted on a coloured ground, and that 
 ground should be a cold one, otherwise it might catch the eye first; any tint 
 of grey down to black will do: but both one and the other should be mellowed 
 by a mixture of yellow, to soften down that extreme cold hue they would 
 otherwise have, which would be unpleasing to the eye. 
 
 The contrast,or balance to white is, as has been said, grey and black. 
 The harmonising tint, yellow, from its full to its palest gradation; or,if the 
 subject will not admit of it, then a yellow green, from its full to its palest 
 gradation ; and if the subject will not admit of this in the latter state, then 
 the painter’s ingenuity is called upon, to introduce one or both these tints by 
 some other means, in this and every other similar instance, but with care not to 
 offend against propriety in the subject. The reflected tint may be composed 
 of some of the principal warm,and next harmonising tints, deadened by black, 
 and made paler with white if requisite. The reflected tint is meant to express 
 that warm lighter shade, to be seen on the dark side of all round bodies at the 
 
16 
 
 extremities, and is supposed to arise from some other object placed in the full 
 light, near to the shadowed part of that you are painting, upon which it 
 throws a reflected light on its extremities : as will be evident, if two balls aie 
 held near each other in a line with the light; take away the furthest from the 
 light, and no such reflected tint will appear—all will be darkness on the shade 
 side of the other, if near no other object, and it will stick to the ground if it 
 happen to be as dark as the shade; but produce the reflected tint again, and 
 it will appear to start from it, or from your paper by the same management, 
 in proportion to the judicious arrangement of your tints. It often happens, that 
 the stalks, and green leaves belonging to white flowers, have a tint of other 
 colours, as yellow, orange, red, &c. ; in that case it must be remembered, 
 that they can only be admitted in their fuller pure state, in the very smallest 
 proportion; and that if there is necessarily a larger portion of them, they 
 must either be flung into the shade, or their strength broken by another tint; 
 otherwise, it would be a mixed composition, and not a white one. White 
 objects that have an inclination to any other colour, either in parts, or alto¬ 
 gether, lose their place in composition as white ones, and belong to that class 
 of colours they have a tint of, and may be considered as lighter reds, 
 blue, &c. 
 
 Note. There is another cause for the extremities on the shade side being 
 made less dark than the deepest shade, setting reflection out of the question; 
 and that is, that in point of perspective they should be so, as objects weaken 
 in strength of colour as they retire from the eye, both in the light, and in 
 the shade. 
 
17 
 
 COMPOSITION OF YELLOW. 
 
 The contrasting,or balancing tint to full yellow, is purple in its deepest 
 degree ; to which such a degree of black must be added as will not destroy 
 the purple : for though purple is the most opposite to yellow, yet place purple 
 in the shade, and it will there appear of a deeper hue than in the light; and as 
 the balancing tints are always placed in that part of a composition that is 
 most in shade, it is therefore necessary to add black to express the effect that 
 shade has upon this, and all other colours in that situation. Its harmonising 
 tints are orange, and pale yellow green; the first, being neaily equal to 
 yellow in lightness, harmonises remarkably well: care must be taken, as in 
 the foregoing composition, not to let the orange be too predominant, and to 
 observe the same rule in regard to the reflected tints. If orange is admitted, 
 a tint of blue should likewise be visible in the shade tints: but it may so 
 happen that orange cannot be introduced; in that case, green, in its different 
 gradations, forms the harmonising tint. 
 
 A pale yellow composition is managed exactly in the same manner, only 
 weakening the contrasting, harmonising, and reflected tints, in proportion to 
 the paleness of the yellow. 
 
 COMPOSITION OF ORANGE. 
 
 The contrasting or balancing tint to full orange is a blue of the deepest 
 tint, with the aforesaid addition of black. Its harmonising tint is red ; but 
 here, as in the foregoing composition, the pure red must be in the smallest 
 
 D 
 
 
 / 
 
 
18 
 
 proportion, and be made to graduate into the shade, so as not to interfere, 
 but to harmonise with the principfy. As green by no means harmonises with 
 orange stem, and still less with red, you must necessarily introduce yellow, 
 which will give you an opportunity of introducing green agreeably; and,by 
 mixing the orange and blue you^get an olive, that will suit with both the 
 orange and red : you must likewise have some purple to balance the yellow. 
 The reflected tints as before directed. 
 
 A* 
 
 COMPOSITION OF GREEN. 
 
 Nothing pleasing could be made of this as a composition, without 
 taking it in its palest degree, as it graduates from yellow; but by making a 
 jX principal*.of it in that state, there is room to do a great deal, as it has not only 
 nearly the same degree of lightness as yellow, but may be worked into a very 
 rich effect. Its contrasting,or balancing tint ; is pale red, inclining to purple, 
 in proportion as the green inclines to yellow; but this tint must be so subdued 
 by black, as not to be either too bright, or too strong for the green in its 
 palest state. 
 
 The harmonising tints, being deeper greens, titty will also require deeper 
 reds to contrast them; which are to be subdued with black, likewise in the 
 deepest shade. 
 
 The leflected tints are formed, as before mentioned, with the principle ;t- 
 warm tints, and the contrasting one. U 
 
19 
 
 COMPOSITION OF SCARLET. 
 
 The contrast to bright scarlet is green in its deepest degree, with the 
 addition of black aSswiri. The harmonising tint is full red, or deep scarlet. 
 fa ^Care must be taken not to let the scarlet, and pure green,join; as they are too 
 much of extremes to associate in any degree without the interference of red; 
 and even then thejwo colours muj^e broken one into the other,^if orange is 
 introduced, him will^ncrease the harmony; you may then have a 
 
 tint of blue that will assist very much, and if you can introduce a small 
 
 portion of yellow also, it will be an advantage, as you then have a 
 purple tint that will give jtn ^a^ r^ e able effect to tne whole, which, with 
 mrrrly scarlet and green^you li n itt y can 
 
 
 Palejscarlet is managed nearly in the same manner, only weakening, as 
 aforesaid, the balancing, harmonising, and reflected tints; which are by tins 
 means reduced to pale orange, pale red,or pink, with pale lilac,- observing to 
 break, and subdue them,as the occasion requires. 
 
 COMPOSITION OF BLUE. 
 
 This colour, from its cold nature, is so unfit for a princijdf in composi¬ 
 tion, that I only give its management with a reference to its place in a full 
 group.; and even there, it is only by a mixture of white that you can bring 
 it out of the shade, which is its proper place, and where it serves as an 
 harmonising tint to green. Orange is the contrasting tint, but it must be 
 subdued by black, or it would not answer the purpose; and indigo is the only 
 
 D 2 
 
20 
 
 harmonising tint that you can freely make use of; so that it is evident this 
 colour, in its deep state, can never be made a princij^U of. But by a mixture 
 of white you may bring it forward in a certain degree,- and it is in that state 
 I shall speak of it. Its contrasting tint is orange, subdued by black and 
 white, so as neither to be too dark^nor too bright for the blue. 
 
 U 
 
 Its harmonising tint a deeper blue. 
 
 The reflected tint is 
 contrasting tint. 
 
 of Am pale blue, and a small portion of the 
 
 Whatever green yomhave occasion to admit must be carefully subdued 
 by black, to prevent itfbeing too powerful for the blue. 
 
 INDIGO. 
 
 From what has been said of the foregoing colour, it is scarcely necessary 
 to observe, that this colour cannot form a composition of itself. 
 
 COMPOSITION OF VIOLET. 
 
 • • ^ 
 
 This colouijin its deepest state>isjinfit afa© for a principle in composition,* 
 
 and its proper place is in the shade. But as three of the prismatic colours, 
 in their gradations from one to another, form each of them a third for 
 instance, red graduating to yellow produces orange ; and yellow graduating 
 towards blue produces green ; blue graduating towards violet produces in¬ 
 digo so does violet, the extreme of one end of the prismatic spectrum, 
 graduating towards red^/the extreme at the other end, produce in its grada- /4, 
 
 A- 
 
Mju, 21 
 
 , . , full red. I shall therefore consider i 
 
 tions crimson and lost in 
 
 funder four heads J ^^v'li^» crimson, and pin . 
 
 " v ;r b i 
 
 and qualified^- amore conspicuous situation ;-it is then termed lilac. 
 
 The contrasting tint to this is pale yellow, butmust be so weakened,and 
 
 M P d bv Wack and white, as to form a tint neither too deep,nor too 
 subdued by black an tint is a deeper purple: the greens 
 
 Slit tZ Place b> a fell .cap ,,»«!. 
 introduce violet in the shade. 
 
 COMPOSITION OF CRIMSON. 
 
 T „ IS co.oor is 
 
 gU «... fl- Piopec addltioa of black fo, the deep 
 
 shade. 
 
 The harmonising tint is violet. 
 
 The reflected tint, as before, Composed of .be prine^ a„d coatrast- 
 ing tint. 
 
22 
 
 But I shall not give a separate example of this colour, as I shall have an 
 opportunity of introducing it in its proper place, in the following compo¬ 
 sition of pale crimson,or pink. 
 
 The contrasting colour to this is a warm 
 pink, with the proper addition of black. 
 
 green,^jiale in proportion to the 
 
 The^harmonising tint^a deeper pink,or crimson; the reflected tint as 
 before,^ composed of the princijd\ colour, and contrasting tint: if yellow is 
 introduced into this composition it must be carefully managed, so as not to 
 be too powerful; and a little of the^e yellow,and pale pink, .mixed, forms 
 %^^&^ 0mses exc eedingl_yjvith the pink. The yellow 
 
 afe^pale purple, which is also an advantage; andjhay be introduced 
 both in the shade, and at the extremities on the light side. The deep* 
 
 crimson may also be introduced in the shade here, and<ives a more forcible 
 effect to the whole. ^ 
 
 Having now gone through the whole range of prismatic colours, and 
 shewn the management of each colour separately, I shall next attempt to 
 do the same m the arrangement of them all in full groups; .assigning to each 
 tliur proper place and q^mtity, according to thw rwpoottvc stren<nh and 
 power in comp^,t.on£adding such rules as & wi „ be 
 
 necessary- -* —- * - ^ “ 
 
FULL ARRANGEMENT 
 
 OF 
 
 COLOURS. 
 
 In order to arrange a variety of colours in a group, so as to produce harmony, 
 attention must be paid to the order in which they are seen in the Rainbow, 
 or the Prismatic Spectrum*; for,if colours are promiscuously jumbled to¬ 
 gether, without regard to the proper quantity of each, and no other meansjised 
 to obtain harmony than merely subduing them by shade, or breaking them by 
 mixing them with one another, it will frequently be impossible to do it^tlmt 
 way without departing from nature; and,if they could be 
 •OH- in subordination by those means, still the additional attention to the 
 natural lightness, strength , and quantity, of each, is as requisite to be observed, 
 as*Jtfie rules of perspective in regard to the size, and figure of objects: for, 
 
 * It is scarcely necessary to explain the well-known spectacle alluded to: but as there may be 
 some young ladies who have not seen it, I shall beg leave to inform thern.it is a mode of separating 
 the component parts of a ray of the sun’s light by means of a three-corner’d wedge of glass, called a 
 prism, which will fling them upon any object in their way, such as a sheet of paper, and there exhibit 
 exactly 'the same colours^seen in the rainbow, beautifully blending one into another an iixm tilinrr i 
 and formjvhat is called die Prismatic Spectrum. 
 
 / 
 
24 
 
 though an object, drawn contrary to these rules, be s w oi 1 shaded^and 
 
 coloured, so as to escape hurting the eye at first sight, yet, to an attentive 
 observer, the defect would appear, and lessen the value of the picture. So 
 will a group of objects, though separately well painted nenpna^ «£iU 
 
 be less pleasing than when mhmUkmk they are well arranged as to colour. 
 
 
 It may lay a pamtgr under some difficulty to admit only aw o h- ajjuantity 
 of one colour, andjkeep to nature ;—as for instance, in a flower-piece, where 
 
 the -winrk specimen of a red, or any other coloured flower,^placed in the 
 full light,^nay be too much for f|nir >:i ']’ the other colours ;—yet it 
 is by no means an insurmountable one, though it certainly requires judgment, 
 and ingenuity in the disposal ofAegroup. But the exercise of these will 
 be amply repaid by the effect^an observance of the laws of harmony will 
 produce. 
 
 The art of arranging colours lies in bringing all those forward in order , 
 that most nearly resemble light and heat, and putting all those back in order, 
 as they depart from those standards, and apportioning them according to the 
 rule before mentioned*, for the distribution of light and shadow: ob¬ 
 serving that, though one quarter of the globe may be said to be enlightened 
 at one time, yet it is not all equally enlightened; that quarter contains a 
 gradation of light, as does the dark quarter of shade, and the remaining 
 quarters contain a gradation between the two extremes of light and shade: 
 and in the same manner, to produce true harmony, must the colours be 
 arranged, graduating from the extreme bright and warm tints to the extreme 
 cold ones, for if one colour is out of place it destroys that harmony of colour, 
 which is as pleasing to a nice eye, as true harmony in music is to a nice ear. 
 But, though every practitioner in the art of painting must feel the advantage 
 of a scientific arrangement of colours, and all aim at harmony, and succeed in 
 a degree proportionate to the method taken to produce it, yet no one can 
 succeed entirely without a fixed principle, or theory to work by; but must 
 
 In a preceding page of this work, and which will be found to agree pretty nearly with what 
 Sir Joshua Reynolds has said upon the subject. 
 
25 
 
 frequently be at a loss, as I confess myself to have been, from want of clear 
 ideas on the subject; and the conviction that those must be obtained before 
 perfection could be arrived at, has been like a weight upon my pencil, pre¬ 
 venting its exertion, making me work without confidence, or satisfaction, and 
 constantly in quest of the guide I wanted. 
 
 Lairesse proposes a method of coming at the point, by making use 
 of coloured patches of silk, or paper, and placing them before the eye in 
 gradual succession, giving precedence in order to those which struck the eye 
 first, and so on. And his method might answer, in a certain degree; but, 
 in many instances, it would leave the eye doubtful and undecided, and give 
 no information to the judgment, why such and such tints should be preferred, 
 or placed in such, or such situations. 
 
 But some late philosophical experiments appear to afford a surer test, as 
 far as they go, and promise, I think, a certain principle to act by . 
 The experiments I allude to, are those by Dr. Herschel, for ascertaining 
 the illuminating, and heating power of the rays of the sun : those on he 
 power of illumination answer, in fact, to what Lairesse practised with the 
 coloured patches, only they were tried in anicer manner with the microscope, 
 than could bfeh the naked eye ; andj^so exactly specified as to save 
 any one else tCe trouble of repeating the experiments. They deteimine the 
 situation of the colours in regard^ each other; but the quantity o eac i 
 must either be determined by tfegr respective heat, or strength, oi by the 
 prismatic proportions; possibly they may correspond: for though Doctor 
 Herschel does not specify all the experiments lurried as to heat, he says 
 
 * I have sought assiduously from books, &c. and have received information on many different 
 
 points from tire practice of others; yet I have never met with any entire system,or theory for the 
 points ,fro P h i ^ ^ oU masters tQ0k the rambow for their guide, the 
 
 Importunity I have had of seeing their works, has prevented Reserving how they availed /*** 
 themselves of thl lessons it afforded them. They also possessed a degree of philosophical know g , 
 
 rn.de tl«« profit — ■I"'”! ” T 
 
 it to advantage. 
 
 E 
 
36 
 
 jhffhz ^traced it through the whole ^prismatic range; that it begins in the 
 invisible rays, before it reaches the red, from whence it extends to the utmost 
 limits of the violet rays, and that it is gradually impaired as the rays grow 
 more refrangible; by which it appears, that it is not equally distributed 
 among them, hut is chiefly resident in the red rays, and may be proportionally 
 so in all: however, till that is fully ascertained, we cannot judge whether 
 it would afford us a better guide, or not. But it appears to me, that if the 
 relative degree^oj' heat, as well as illuminating power, which colour bears 
 to another, were exactly known, a scale might be formed, shewing the proper 
 accompanying tints, both in a pure and compound state, to whatever principle 
 one was fixed upon for the centre of a group : till then, we must make use of 
 the best guide we have, which is certainly that of the prismatic proportions, 
 0t. flta- <)r< ^c^hc^stam^«, as to degree of illumination; and possibly none will 
 
 ever be iffiwUJ thatwill increase the harmony. But though the colours, in 
 a great measure, may be arranged, and proportioned by those rules, the 
 qualities of the colours should be considered, as they all differ in their natures, 
 and produce different effects. There is a strong analogy between them and 
 the note^ip music, for each of them has its separate part to perform in a 
 group, that the different notes have in a piece of music; and till their power, 
 and province are fully considered, a painter has as little chance of producing* 
 harmony in a group of flowers, as a musician, who has not studied the 
 theory of music, and considered the effect each note would have in a full 
 chorus. 
 
 Yellows from its brilliancy, and affinity to light, is fittest to come next it in 
 the light quarter, lied, from its warmth, is fit for a prominent place of 
 power, and force. Orange stands next to it, hut has less power. Blue, from 
 its coldness, is fittest for a retiring place in the shade quarter. Indigo, from 
 its affinity to black, is fit for the deepest shade. Violet, from its weakness, is 
 only fit for a retiring situation. Green has a double quality; it is equal to 
 yellow m point of illumination, comes next it in the light, and yet from its 
 cold nature is equally fit for the shadow. But though blue, indigo, and violet, 
 are capable of distinct illumination, and blue equally with red’, not one of them 
 
has warmth 
 colours, or 
 conspicuous 
 
 The following shews the Prismatic order, and proportion of Colours. 
 
 Red 
 
 - 
 
 45 degreei 
 
 Orange - 
 
 - 
 
 27 
 
 Yellow - 
 
 - 
 
 48 
 
 Green - 
 
 - 
 
 60 
 
 Blue 
 
 - 
 
 60 
 
 Indigo - 
 
 - 
 
 40 
 
 Violet - 
 
 - 
 
 80 
 
 enough to fit it for a prominent situation in the mass of warm 
 as a°princip!f one by itself, but may be introduced in less 
 f- 
 
 places. 
 
 Th, follming shewn the order they stand in, in point of illumination, as proved 
 from the before-mentioned experiments. 
 
 The highest degree of illumination lies between 
 Bright Yellow, and 
 Pale Green; 
 next Orange, 
 then Red, 
 
 and Blue equally with Red ; 
 then Green, 
 
 Indigo, 
 
 Violet. 
 
 The result of the foregoing is, that though the prismatic order of' co¬ 
 lon,s must, in some degree, guide their arrangement ,n a group, 
 
 E 2 
 
•V\ X 
 
 K 
 
 -I 
 
 28 
 
 must be deranged to suit the order of them in point of illumination ; for it 
 is by the latter rule a painter should dispose them in a picture : but still 
 there is a difficulty, owing to red and blue coming next one another, as 
 being equal in brilliancy , and yet they are such opposite colours they never 
 will harmonise together, but require the intervention of another colour. 
 Green would answer the purpose, but I prefer inverting the order of them, 
 and placing violet next the red, which will, on the whole, have a better 
 effect; for green coining last, serves, in some measure, as a balance to 
 yellow, the first and principal colour after white; and the violet being 
 blended * with the red, and partaking of its brilliancy and warmth, an¬ 
 swers better, in point of illumination, than the mixture of red with either 
 blue or green. 
 
 This alteration is simply no more than joining the two extremes of the 
 nsmatic Spectrum, which, from the almost opposite degrees of strength 
 in red and violet, seems a contradiction to what is right; but we shall 
 
 thus-— 11 W ' U ‘ tnS< ' er “ Factice ’ and h } T this arrangement they will stand 
 
 Yellow - - 
 Orange 
 Red - 45 
 
 f Violet - - - 80 
 
 Indigo ... 4Q 
 
 Blue - - - 6 o 
 
 Green - - - 60 
 
 To bring this into use, draw a line upon paper, any size you choose, and 
 
 * It must be remembered, that all the prismatic colours blend gradually one into another, and 
 by that means form intermediate tints. ’ 
 
 . 1 l W ° Uld ^ Sh t0 be *at I consider my own opinion of this arrangement as deci¬ 
 
 sive, but as a matter that requires the investigation,and sanction of other practitioners in the art j though 
 
 Srto red ^ ^ ° f ** C ° loUred « the violet sid°e. 
 
 - MUM - 
 

 fl 
 
 r> ZZ Ztr.t aZletzeZy Zeen *taicZ fZ/at it u nut tt/tirt// tit atta:'tiy& tZit coZotiM Zn. cnlcZ*) ty 
 ZjZ/.ctu/<e;tZt aZtuiuZtZy of t/uit. yua.-rvtity n/earAt ctnZiZ yutt. At 
 
 f'Au / tt/Z It.) /titty At ,//(*’// AeHe. tyAytic/i r/?rA //>//,>y'/ZZiZ /iw/Z /At otttn,cttZttttf'. ZyZutieyZiZZ 
 
 'tuit/uny mole ‘ttut^t■/& ZetttZetZf&l/ vn>t/uj tZutm tAteac/t cttAntr yn , 
 /■t.^/rt-/ /n tAe o/Zir '/d, tit/ttt/t m *t/ieums Zy tZu .j't/tu //y'/if l/tnn tAat~ m ettAtnt/ieiZ, aniZ yZ^ZZ 
 
 ty ZZtZ/ it //'A czrzizztf/tjzz zz/y£zz£ z*z cz??zjzz/eiez/, z/' ?r? am/ />az /nozvzi 
 
 e^€c£ ISaa/ZaZ Z* Z/uza/c 07< 
 
 eZt'lectZy. uZuitcZylee o/ZZiiZZcoMoyol itAat/t ^jAouZZ Ae yoi/tnii/ OretzuA rttZ'Wt/ /, 
 'Z'^Z/Zi/'nt/utity t/tZ> -ZZAtj' ZZtnnnZZ 'ZZ t/tt-ir t/Z/ty An /tu,/uZ utAe'ne-ZZZae'MAt/ 
 /trim, /m/t/it . 
 
 /t ft'/ 
 tUi 
 
 >y 
 
 ZoyJ,/., 
 
 'T 
 
 JhtbUshal as ihcAcb directs. Nov r i.1804, by T. Oardiner, Frinces Street, Cavendish Square. 
 
29 
 
 divide it into 3£>0 parts, or degrees ; set one foot of your compasses in the 
 centre, extend the other so as to take in 48 degrees, that is, 24 on each side 
 the centre. On the same centre extend it again, so as to take in the proper 
 portion of orange, do the same with red, and so on with the rest of the 
 colours, filling each circle with the colour belonging to it, and blending 
 each into the contiguous ones. Thus you will perceive all the colours 
 arranged as it were on the surface of a ball *, in a sort of perspective suc¬ 
 cession, each coming forwarder retiring from the eye according to its na¬ 
 tural brilliancy : and if any one intrudes upon the eye, before its superior, 
 you may be sure there is a fault in its situation. But if these colours were 
 arranged in this order, in a picture, I believe there will not be found any 
 greater error than what the necessary introduction of light and shade would 
 sufficiently reduce to harmony, provided a nice attention is paid to the 
 choice of the tints we make use of to imitate those in the Prismatic Spectrum : 
 for it must be considered, that the pigments we have to express these colours, 
 are far inferior in brilliancy—we have not any, I believe, equal to them. As 
 to the red, vermillion will not express it, neither will red-lead, carmine, nor 
 lake separately but those that appear to me the nearest to 
 
 all, will be mentioned hereafter. 
 
 It will be said, cplours arranged in circles in a picture would be very 
 ridiculous, which it certainly would; but that is not meant: it is only in¬ 
 tended to shew the proper situation for objects of such and such colours 
 in respect to the others, not that there should be entire circles of them. If 
 you admit white into your picture, in its pure state, it must come before 
 yellow, and a space be left for it: as to the proportion, or quantity of it, 
 that will be decided by the portion of pure light (or •white to express it) 
 supposed to be on a ball, and by the same rule in your picture; so that the 
 size of your picture must be your guide. The portion of each of the other 
 colours must be regulated by the proportion they bear to one another in the 
 Prismatic Spectrum, which is sometimes a difficulty, when the objects you 
 
 * If they were thus arranged on the surface of a real ball, their perspective situations would be 
 more readily perceived, as the prominence of the ball would really give each its proper distance. 
 
 
30 
 
 mean to represent do not contain the proper quantity of each separate co¬ 
 lour ; for instance, you may have a yellow flower that contains more than 
 the portion of yellow you ought to admit into the group, and the question 
 is, what to do with it, and not destroy the character j or beauty of the flowei 
 by leaving part out. Here ingenuity must be called to your aid, for no one 
 colour should predominate improperly over the others, but all be kept in 
 their proper force as well as station. When I speak of this piopoition of 
 each colour, I include all the variety* of tints, from the pure to the com¬ 
 pound state : for when the proper quantity of light and shade is admitted 
 into the picture, a very small proportion of the colours will remain in their 
 pure original state; the greater part will be converted into compound light 
 and compound shade tints. Besides this effect that light and shade will have 
 on the general mass, it must be remembered, that every one of the prismatic 
 divisions contains a gradation from a pure colour to a compound one with the 
 next. The 45 degrees red contain pure red, also orange red in all its gra¬ 
 dations, as do all the other divisions their gradations also ; and if this 
 Spectrum is attentively observed, it will soon be seen how small a portion of 
 pure colour there is in each division, that is, in each of the coloured pencils 
 of a ray of light, though they go by the general titles of yellow, orange, red, 
 &c. &c. and must be admitted into a picture, by the very same; yet, if those 
 separate gradations are not nicely copied, and the portion of them all together 
 regulated by the proportion they bear t o^one another in the Spectrum, their 
 effect in a picture will be very different Tawhat it is there. I do not mean 
 that there should be so many degrees of red, &c. &c. exactly as in the 
 Spectrum, for it would be impossible to measure out colours that w'ay, but 
 it will be easy to observe what proportions they bear to one another, and to 
 let the colours in a picture have the same: for instance, there are 48 degrees 
 of yellow to 27 of orange, that is, nearly double the quantity of yellow' to 
 that of orange; therefore there should be the same proportion of them in a 
 picture, in all the different gradations observable in the Spectrum : and the 
 greatest nicety, in this, must be observed, for fear of destroying the balance 
 
 * Which comprehends an infinite variety of other compound tints, formed by a mixture of the 
 different prismatic colours with one another, that will take their station in the picture in point of 
 brilliancy, according to the affinity each bears to that particular pure tint it partakes most of. 
 
31 
 
 of the warm and cold colours, the latter of which, if summed up, bear just 
 a double proportion to the warm, reckoning green in ; for there are of the cold 
 240, of the warm 120 degrees; and, I believe, the nearer this proportion is 
 kept to, in a picture, the more harmonious it will be. 
 
 It sometimes happens that a picture consists of several different masses ; 
 in that case, the pure prismatic colours, with their several compound light 
 and shade tints, should occupy the first mass. The second is composed of 
 double compound tints; the third, of still more obscure ones : a table for 
 forming all which I shall add at the conclusion. When a picture is com¬ 
 posed of only one mass, all the colours, both pure and compound, in their 
 several degrees, may also^jje introduced; and it will, in either case, have all 
 the force and effect that Js in the power of colours to give. But I must 
 here observe, that a picture composed of one mass, will never please the 
 eye, in point of contour, so well as one that consists of several different 
 ones ;—for distance* always lends beauty to the view : therefore, if obliged to 
 have only one, I should, if possible, contrive to break it so as to produce 
 the above effect as far as I could. 
 
 It frequently happens that a picture is composed of objects that have not 
 one pure tint in them ; in that case, they cannot have the same brilliancy, or 
 force of effect; they are unavoidably more obscure, that is, less striking; 
 but still may be in perfect harmony, if arranged by the same rule as the pure 
 colours, and in the same proportion as aforesaid, but they will never come 
 so forward to the eye; the strongest effect of light and shade and contrast 
 altogether, will not force the compound tints so forward as the same degree 
 of light and shade contrast will the pure unmixed prismatic ones, which 
 plainly shews the latter are fittest for a prominent place, when you are at 
 liberty to introduce a full assemblage of them. Prom the foregoing it will 
 appear, that to form a good picture is not the work of mere fancy, there 
 
 * “ ’T is distance lends inchantment to the view, 
 
 “ And robes the mountain in its azure hue.” 
 
 Pleasures of Hope. 
 
! 
 
 requires a great deal of thought and knowledge to produce one, even though 
 it consist of nothing but flowers. 
 
 The first thing to be done at the commencement of a picture, is to con¬ 
 sider what your materials consist of, and what effect you can make them 
 produce : it is not often that a painter has all he wishes; ingenuity must 
 often make up for the want of some, and the imperfection of others; and 
 it is no small difficulty sometimes, when a strict attention to nature is re¬ 
 quisite. But the best way to get over the difficulty, after you have made 
 your drawing, and considered what parts you mean to come forward, and 
 ijrhTr retire, is, to make a blot, of such colours as your subject allows 
 you to introduce. If the whole range of colours is allowed you, dispose 
 them as has been already mentioned, placing a prominent colour on a pro¬ 
 minent part, and the retiring ones in their respective situations. If you 
 have only part of the pure and brilliant colours at your disposal, then you 
 must lessen the quantity of the other colours, rejecting the contrasting tints 
 to those you leave out. If there be no pure yellow in it, there wants no pure 
 purple; if no pure orange, then no pure blue is necessary: but if obliged to 
 have pure blue, &c. in short, to form a picture out of discordant colours, 
 then is the time to exert ingenuity, and by some means or other to har¬ 
 monise them together, which is seldom impossible in works of fancy. 
 
 I must now speak a little further upon the compound colours, lest I 
 should be misunderstood in what I have said respecting their situation in a 
 picture. We must first consider which are the compound colours; and it 
 appears that the Prismatic Spectrum itself presents you with an alternate 
 range of pure and compound* colours, by the blending of one with another ; 
 
 * At least apparently so, though there are some who consider them all as elementary ones; 
 whether they are so, or not, I cannot take upon me to say. But I cannot help considering violet as 
 one, instead of indigo, and the latter as a compound; and I am inclined to think so the more, because 
 a pure and a compound tint alternately are visible, in the prismatic range, as low as blue : and it 
 does not seem probable, to me, that their order should change there, and for which there appears no 
 other cause, but that indigo and red (it is said) will produce violet, if it may be called one, for it is 
 many degrees inferior to the real one; whereas violet and blue will produce a much nearer imitation 
 of the prismatic indigo, therefore, in that respect, I think there is stronger reason to consider it as a 
 compound colour than violet. 
 

 33 
 
 for what is orange, green, and indigo, but a compound each of the two 
 colours contiguous to them'?—these may he termed pure compounds; also an 
 equal mixture of any of the remaining colours : but when light or shade, 
 that is, white or black, is mixed with them, then they may be deemed 
 treble compounds ; but, mix three of them together, and you produce a more 
 undecided colour,—for instance, red, orange, yellow, and the mixture will 
 be neither a true red, a true orange, nor a true yellow, the just balance that 
 formed the red and the yellow into a pure orange is destroyed, and you get 
 a compound that is neither the one nor the other: or take yellow, green, 
 blue, and again the balance is disturbed; it is not the pure green, but a 
 gradation from it towards yellow: the same with blue, indigo, and violet; 
 it becomes neither a blue, an indigo, nor a violet, but an obscuier tint dif¬ 
 fering from all. The foregoing, it must be observed, aie successive piismatic 
 tints, and it is only the varying from the prismatic proportions, in some of 
 the instances, that produces the difference mentioned ; but when you come 
 to mix tints not contiguous to one another, you get into an obscurer class : 
 for instance, red and violet; red and indigo; red and green ; orange and 
 violet; orange and blue; orange and green, &c. &c. Now, I believe, it 
 will appear in some degree from this, bow necessary it is for every one who 
 wishes to excel in painting to be rvell acquainted with the mixture of 
 colours, that is, the variety of different tints that may be formed by that 
 means; and that it is working quite in the dark, ami leaving it to mere 
 chance, to attempt producing any particular effect, till the means of doing 
 it are fully understood;—that, without knowing it, is proceeding pretty much 
 upon the same footing with a person Avho sits down to copy a Greek^or an 
 Hebrew manuscript, ignorant of those languages ',—the chaiacters may be 
 exactly and beautifully imitated , still the sense is lost: but (to carry on the 
 allusion) if the languages are understood, then the meaning, and beauty of 
 the subject are as it were translated into your mind, and another such sub¬ 
 ject, or another such picture, may be formed from the information gained ; 
 you then work as an original performer, not as a mere copyist. 
 
 
 _ _ 
 
 _L 
 
34 
 
 I come now to shadows, which I omitted speaking of in the early part 
 of this work, intending to speak of them here, and consider them both as 
 to form and colour: a very few examples will serve to shew^they must be 
 of whatever form the object is that^hrows them, and that they are subject 
 to tbe same rules of perspective fcbffit the objects themselves abb.; but for a 
 nicer delineation, a fuller knowledge of that branch .of perspective is ne¬ 
 cessary : and, though it may seem foreign to fl^werrdrawing, I must beg leave 
 to observe, that it matters not whether the .asMaiaw a shadow is projected 
 upon is hard or soft, if of substance enough to receive a shadow upon it; 
 and that it will always add beauty to a picture to have the shadows thrown 
 with truth and judgment. 
 
 In regard to the colour of shadows, here observation must be called in 
 to aid, though there are some general rules to be followed as guides; for 
 instance, that two contiguous colours mixed together will form the colour 
 of the shadow, projected from one upon the other, strong in proportion to 
 the degree of light upon it, and obscure or indistinct in proportion to the 
 opacity of the object that throws it, which may be such as to render the 
 interior part of the shadow quite dark, and only visibly coloured at the 
 termination; or the object may be so transparent as to cause a very slight, 
 shadow, in which case, the colour of it will also be slight: but expe¬ 
 rience must supply nicer rules. I shall only say further, that the putting- 
 in the shadows in their proper tone, has a very great share in the harmony 
 of the picture, and that this alone shews the necessity of studying the com¬ 
 pound tints. 
 
 I hope I shall not, by what I have said, be thought to have put an 
 undue restraint upon placing the colours, or not to have left a practitioner 
 at full liberty, who certainly may make choice of whatever colours are 
 agreeable, and have a picture either brilliant or obscure. What I wish to 
 enfoice is, that when that point is determined, that p 7 'oper colours may be 
 employed to produce the desired effect, and not to expect it otherwise, for 
 some parts would either be too warm,or too cold ; in short, it would never 
 be in harmony, if the perspective gradation of the colours (if I may use 
 
that term) is not observed : but I by no means insist that 1 have succeeded 
 in it, for some change may be found necessary, on trial, that my limited 
 practice has not enabled me to make ; therefore, what I offer is with entire 
 submission to future experience, whether it arise from my own practice, 
 or that of others, though with some confidence as to the principles I have gone 
 upon, which I believe are a sure guide to form a theory upon, though others 
 may succeed better than I have done ; and should these pages fall into such 
 hands, I must beg to be understood that I presume not to offer them to 
 my fellow artists, but only to those pupils whom it is my lot, and my duty 
 to instruct to the best of my power; and which a desire of doing more fully 
 than the space of a short visit has sometimes enabled me to do as I wished, 
 has been one cause, with other considerations, for making them public ; 
 and my intention, I hope, will plead my apology for many imperfections, 
 no doubt, there are in them, that I am unable at present to discover. 
 It is scarcely necessary to add, that in speaking, as I have done, of the 
 arrangement of colours in perspective succession, I mean as to the general 
 mass ; and that all the colours, warm, cold, and compound, may be 
 interspersed throughout the whole, independent of that, provided the ge¬ 
 neral effect, or mass is not broken, or so much interrupted as to render it 
 indistinct, and -not striking at the first view. A thousand lesser beauties 
 may be introduced this way, that must entirely depend on the painter’s taste 
 and judgment. 
 
 I have endeavoured to shew, in the foregoing pages, the disposition of the 
 Colours necessary to be observed in the objects represented, but I have often 
 been at a loss, in my own practice, how to arrange them in the back-ground; 
 to an ignorant person it may seem of little consequence, but I am not singulai 
 in saying, it requires full as much judgment, to do it properly, as any part 
 of the picture*; it depends entirely on the colours in the object you re¬ 
 present, and must vary with them, keeping to one certain rule, to observe 
 
 * Rubens was asked by a friend to take a young man under his care, and, as an inducement, was 
 told he could already paint, and could paint him his back-grounds : Rubens replied,. “ If he can paint 
 my back-grounds, he does not want my instruction in the art.” 
 
 F 2 
 
36 
 
 the same tone throughout, and vet that it shall be so neutral as, though it 
 makes part of the picture, yet^cioes not interfere ; or intrude upon the eye 
 before any one parlor object in the group, but that it is fairly behind them 
 all, at whatever distance you choose it to appear; whether it be one even flat 
 surface ; or a gradation of distant objects, still it must be in union with the 
 whole, and the general colour of it formed by a mixture of the separate ones 
 in your group, only not in the pure, but an obscure*, compound state. If it 
 consists of a dark sideband a light one, the one should oppose the light side of 
 the group, the other the dark side of the same; and the same rule should be 
 observed in your fore-ground, only increasing in brilliancy, but still keeping 
 them so neutral as that they do not rise above the surface they were meant to 
 represent, consequently that no pure superior tint is to be employed there, fi/KOL. 
 those can only occupy a single part of the picture; if otherwise, it is the 
 same as if there were twojor more equal lights in it, which never can be if you 
 keep to nature. 
 
 In a preceding page of this work I have mentioned, that an agreeable 
 picture might be formed,'in many instances, with one principl*—a harmonising 
 and a contrasting colour: but I said nothing of the quantity 5 or proportion of 
 each—-that required more explanation than could be given in that place with¬ 
 out interfering with this succeeding part; but the same proportional- quantity 
 of each must be observed with a limited number of the colours^ that you 
 observe with the whole range, and these with the compound tints; that the 
 admission of light and shade, and mixture with one another, produced in the 
 shadows and reflected tints, may be worked up into a good effect. 
 
 The prismatic range, as before observed, presents an alternate change of 
 pure and compound tints, as may be found by your being able to produce 
 the latter by a mixture of the two colours contiguous to the second, fourth, 
 and sixth; in the same proportions they exist in the Spectrum, and may be 
 called pure compounds : the change in the order of the colours, with a view to 
 give precedency to some on the score of brilliancy, must be recollected, and 
 
 * Throughout this little work I must be understood to allude only to flower-pieces. 
 
37 
 
 that it will make a difference in regard to the compounds, which, under that 
 change, will not all fall alternately in that arrangement*but that is not ma¬ 
 terial to their situation in a picture. I scarcely need repeat that these colours 
 belong to the first mass, or most prominent part of a picture. 
 
 I come now to those fit for a secondary mass, the most brilliant of which 
 are formed as follows, and form the first link in the gradation from pure to 
 obscure colours. It has been shewn, that two colours contiguous to an in¬ 
 termediate one, will form an imitation of that colour; so will a mixture of 
 two compound colours produce an imitation of a pure prismatic r one, thoug 1 
 verv far inferior in brilliancy, but still not the less useful to a painter. Again, 
 these secondary pure colours, that is, yellow, red, violet, and blue, will orm, 
 by mixture, secondary compound ones, orange, indigo, and green; and these 
 secondary compounds will form, by mixture, pure tints of a third order and 
 these a-ain compounds as before; so that supposing a picture consists of 
 three distances, you have a distinct set of colours for each, the formation of 
 which is set down in the following tables, and they may all be made lighter or 
 darker by the addition of white or black. An infinite number of other 
 compound tints may be formed out of these three different tables, by varying 
 the mixtures and the proportions; but to give tables of them also would lead 
 me into so wide a field that it would take me beyond the bounds I prescribed 
 to myself, and further, perhaps, than those into whose hands this wr 
 would be inclined to follow me. But should any one choose to pursue these 
 trials I refer them to Mr. Gal ton’s Experiments on Colours, and to Mr. 
 Harris’s System of Colours : in the latter they will see the whole range of pure, 
 and compound colours, and the contrasting tints to each, at one view. 
 
 * Though even there every intermediate tint may be formed by the two contiguous ones. 
 
 \im 
 
!'•: f 
 
 
 
 
39 
 
 TABLE I. 
 
 Elementary ones. 
 
 48 
 
 Yellow 
 
 45 
 
 Red 
 
 80 
 
 Violet 
 
 Prismatic Colours. 
 
 Compounds. 
 
 48 Yellow 
 45 Red 
 
 compose Orange 
 
 80 Violet 
 6o Blue 
 
 compose Indigo 
 
 48 Yellow 
 6o Blue 
 
 compose Green 
 
 60 
 Blue 
 
 These colours, with the addition of White, to lighten, and Black, to darken, 
 belong to the first mass, or most prominent part of a group, and have no place 
 in any other part of it. 
 
43 
 
 TABLE III. 
 
 Compound tints of the third order are not set down, it being unnecessary , 
 as they are formed from Table II, exactly in the same manner as those are 
 from Table I; and produce similar tints, only weaker. 
 
 I shall only add, that in mixing these colours, the imperfections of the 
 pigments -we have to use must be allowed for;'for they will all fall very far 
 short of those seen in the Prismatic Spectrum; but I believe the following list 
 will come as near as any: 
 
 Gamboge 
 
 Red-lead and Gamboge 
 Vermillion and Red-lead 
 Lake and Prussian Blue 
 Prussian Blue 
 Antwerp Blue 
 
 Antwerp Blue and Gamboge 
 
 for Yellow 
 for Orange 
 for Red 
 for Violet 
 for Indigo 
 for Blue 
 for Green 
 
 
APPLICATION 
 
 Having, I hope, rendered my meaning clearer to phdosophical readers 
 in the foregoing pages, than it was in the first edition of this Essay, I 
 wish to make the work more useful to those also who wish for a guide 
 to their Pencils, in forming groups of flowers, by giving particular explana¬ 
 tions of the blots. 
 
 I shall therefore suppose, for sake of illustration, that each blot is a 
 group of flowers; but must at the same time observe, that they have not 
 been formed with the most distant idea of their being examples, in re¬ 
 spect to the contour of flowers. They are merely compact blots of colours, 
 exhibiting the effect produced by arranging them according to the theory 
 delivered 0 by the foregoing pages; and as such, I shall endeavour to explain 
 the principles of their formation, by which they may be made 
 the student in painting. 
 
 iMMI 
 
 ■ wm 
 
46 
 
 If I was going to form a group of white flowers (say white wild roses), 
 I would place my principal flower at No. 1, the most prominent part of 
 the group, and of course most opposed to the light; and there, white should 
 reign in full power, and in such a proportion as the size of the group (not 
 the blot) requires. 
 
 I should place other flowers at 2 and 3, but the white in them should 
 be so subdued, that the eye would not be caught, but conducted by them 
 to the principal flower; the intermediate spaces between them and it 
 should be filled with pale green leaves of a yellowish hue, so as to har¬ 
 monize with the whites; and the strength of the colour between the full light, 
 and the dark side, should be increased to a full tint; then gradually joined 
 and mixed with the contrasting tint, which of course should be placed in 
 the deepest shade. 
 
 But a question will occur, by what means shall black, the contrasting 
 tint to white in the petals; or purple, the contrasting tint to yellow in 
 the seeds; or red, to the green of the leaves, be introduced? for these 
 three several colours, white, yellow, and green, all require their contrasts. 
 
 l\o subject could perhaps be found that can so easily supply them 
 all, as a group of wh.te roses; for in the leaves, buds, and stalks of the 
 w ite og rose there are all the above tints in a variety of shades; so 
 that the aid of no other object, nor flower, need be called in. These 
 co ours therefore may be dispersed throughout the shaded side, in as deep 
 a degree as is wanted; and also throughout the whole in a weaker de¬ 
 gree, so as to break the pure tints in all the minor shades, and give 
 value to those in the prominent parts of the group; for neither the whites 
 
 Lethe ’ 7 6r tbtS ^ the IeaV6S and ^ must be kept 
 
 together in one place; but each should be distributed throughout the whole 
 
 one clas Jr 2 0Slt,0 \° f 7?' be fou " d the P-st degree of 
 
 them m the highest lights, and of the other in the deepest 
 
 
47 
 
 shade; and if gently lightened towards the extremity, so as to form the 
 reflected tint on the dark side, it will at once give roundness to the whole. 
 
 By this management, that breadth of light and shade may be accom¬ 
 plished, which, whether the group be highly finished, or not, cannot fail of 
 having a good effect. 
 
 It being recommended to the student, to employ only three colours at 
 first, to form a group or composition with, as exemplified in these blots, it is 
 requisite here to observe, that the introduction of more is not a matter of 
 choice, but of necessity, arising from the nature of the subject; for although a 
 white, a yellow, or a red flower, &c. may be chosen for a principal colour, yet it 
 seldom happens but that there is a mixture of other tints in them also; for 
 instance, yellow in the seeds, or red in the leaves, as in the one now chosen. 
 But when that is the case, it is not necessary that there should be the prismatic 
 proportion of each in the group; it will be enough to give theii harmonizing 
 and contrasting tints, in as small a degree dispersed throughout, as the quantity 
 of the yellow, for instance, requires; and if a greater portion is employed, it 
 would become an example of four, or as many colours as are introduced, which 
 would render it more difficult, and not in that instance so true a delineation 
 of nature. 
 
48 
 
 APPLICATION OF THE YELLOW BLOT TO A GROUP OF 
 
 YELLOW FLOWERS. 
 
 Under the idea that each of these blots are a group of flowers, I shall go 
 on to remark, that yellow roses are chosen to exemplify the one before us, 
 because these flowers differ entirely from the white one, in one respect; namely, 
 that it affords no contrasting tint within itself; the plant being painted by na¬ 
 ture, as it were, with two colours only, yellow and green, except, indeed, some 
 brown in the stem and branches: more skill therefore is required to form a 
 pleasing group of these flowers, than of the foregoing one of white roses; 
 especially as the green leaves of the yellow rose are of too blue a tint to harmonize 
 with the brilliant yellow in the flower, when placed in the full light at A. As 
 a cold tint, in any quantity at least, ought not to be admitted there, none 
 of the green leaves, therefore, should be placed near the brightest yellow, 
 but some more of the flowers should be so disposed as not to receive the 
 light fully on them at B and C, by which means a second degree or shade 
 of yellow may be obtained, and intervene between the bright yellow 
 and the blue green leaves; and thus form an harmonizing tint, either darker 
 or lighter, according to the degree of shade into which they are thrown. 
 Some faded green leaves also may be introduced, which in one stage of their 
 decay offers a variety of beautiful tints, from a full yellow, to a deep brown ; 
 and in producing picturesque effect, every accidental circumstance of this sort 
 may be fairly resorted to, in order to distribute the tints necessary to produce 
 harmony throughout the whole group, when foreign aid is wanted for the 
 purpose. 
 
 The contrasting tint now demands the attention, and as the plant does not 
 afford it, purple must be thrown in by means of some other flower, or object; a 
 ribbon, for example, of that colour twisted round, with grass of a purplish hue, 
 might very well distribute the tint in various gradations throughout the whole 
 
49 
 
 group. The deepest degree of it must be placed at B and JE in the deepest 
 shade, there to oppose the brightest yellow; and if some of it inclined to red, 
 that would oppose the green leaves. But all these colours, both in this and in 
 every other instance of composition, must be very carefully brought in, so that 
 not one of them shall interfere with the principal colour, or exceed the prismatic 
 proportion in quantity. 
 
 It is not necessary, however, to portion out colours in a picture by 
 prismatic degrees, for that is impossible; but it would be very easy to form a 
 scale by them, and for every degree to allow such a certain part, or measure, a 
 square inch for instance, and let that be divided into ten equal parts, as might 
 be convenient for the purpose ; for it matters not how large or how small the 
 scale is: but when once fixed upon, it would be very easy to remember that 
 there should be, for instance, double the quantity of violet to that ot indigo; 
 that there should be equal quantities of blue and green, and the different 
 proportions of the other colours also; in short, that each colour in the gioup, 
 or picture, should bear the same proportion to one another. The effect would 
 then be precisely the same as in the prismatic Spectrum; and for want of this 
 measure, or proportion, being observed, the eye is often hurt in examining 
 paintings (otherwise admirable), where an over quantity of one colour often 
 destroys the value of the rest; and makes one wish, at least, to throw a shade 
 over some part of the exceeding tint, which would, in a certain measure, har¬ 
 monize the whole; but in no degree be equal to that of preserving the true 
 proportion pointed out in the rainbow, which if strictly attended to, would, I 
 may confidently say, produce unvarying harmony in every imitation of it with 
 the pencil. By the blot it may be seen, that some yellow of a much lower tone 
 should be placed near the extremity of the group at D, on the shaded side, 
 in order to produce the reflected tint, without which no roundness or projection 
 will ever be obtained; besides the necessity of distributing the colours generally 
 throughout the whole. 
 
 
50 
 
 APPLICATION OF THE ORANGE BLOT, TO A GROUP OF 
 ORANGE-COLOURED FLOWERS. 
 
 There is no flower, that I recollect, better adapted to exemplify this 
 blot, than the common nasturtium, whose petals display all the gradations 
 of orange, from the lightest shade down to the deepest tint; and some 
 of them also include scarlet, so that there could hardly be found a subject 
 fitter for the purpose; especially, as the outsides of the petals, stalks, and 
 buds, are coloured with a kind of pale broken orange, very favourable 
 for the distribution of the colour throughout the group. The green leaves, 
 too, are of a happy tint, as their yellowish green hue on the upper side, and 
 grey tint on the under, makes it easy to dispose and intermix them with 
 the flower in whatever way fancy may direct. 
 
 In this flower, therefore, we have every thing to be wished for in 
 point of harmony of colour; but in regard to contrast, it is entirely defi¬ 
 cient, and of course blue must be introduced by some other means. If 
 we take a flower to do it, perhaps no better will be found, then some 
 one or other of the convolvolus tribe, which might be beautifully twined 
 and intermixed with the nasturtium: but of these I should not, on this 
 occasion, make choice of the major convolvolus, as in that flower there 
 are two jarring colours, blue and crimson; and the minor convolvolus would 
 have a better effect. The white and yellow in that flower affords an opportunity 
 of introducing both black in opposition to the white, and lilac or purple in op¬ 
 position to the yellow; if the situation of the convolvolus in the group should 
 make these contrasting tints necessary, that is, if the convolvolus has light 
 enough thrown upon it to shew the yellow and white in their pure state, then, 
 the addition of lilac and black in a small proportion, may increase both the 
 harmony and spirit of the group. In the disposal of them in the group, 
 the first thing to be considered, is the proper quantity of each; which, if the 
 
51 
 
 scale has been fixed upon as before recommended, will stand thus; of orange, 
 two parts, seven tenths,* of green, six parts, and of blue, six parts; and then, 
 let the size of each part be what it will, when properly combined, they will 
 produce precisely the same effect as to harmony with that of the rainbow, or 
 prismatic spectrum. 
 
 It is scarcely necessary to observe, that the separate parts or portions 
 of each colour, are not to be confined to one spot or space in the picture, 
 but should be distributed throughout the whole group. Thus at A, I would 
 place some of the orange, where the light is supposed to strike most fully 
 on the group; and the rest I would distribute throughout, some in the 
 shade, and some at the extremities, without apparent design. 
 
 The darkest shade of blue, and some of the darker green leaves, should 
 be disposed of at B; and some of the orange, mixed with these two colours, 
 would form a reflected tint at C, and complete the dark side of the group; 
 while lighter shades of these colours might be thrown into ail the surrounding 
 parts, by means of the flowers and leaves being placed in different positions, 
 as fancy would direct, although some little study at first would be requisite, 
 before the management of the colours became easy. The eye would 
 soon become accustomed to it, and there would be no occasion to have re¬ 
 course to formal measure, which would only be necessary, till the eye was 
 capable of judging without its assistance. 
 
 * Of an inch, or half a one, or of any space or compass that is suitable to the purpose; 
 for the size of the scale must of course depend on the size of the picture and number of colours 
 introduced. 
 
52 
 
 APPLICATION OF THE SCARLET BLOT, TO A GROUP OF 
 
 SCARLET FLOWERS. 
 
 The pencil of nature seems to have been more frequently dipped in 
 this colour, to beautify the various flowers and plants, &c. that display 
 the numberless shades and varieties of its tints, than almost any othei colour. 
 
 The whole tribe of poppies, anenomes, ranunculus, tulips, with 
 many other beautiful specimens of the varieties of this colour, afford ample 
 choice of subject, to exercise both the fancy and judgment ot the painter, 
 in disposing it in a group; but amongst this variety, there requires some 
 care to select the shade of colour that may be denominated pure scailet, 
 or prismatic red. I believe the poppy will aflord it more exactly than any 
 other flower; therefore we will suppose the present blot to be a group of 
 scarlet poppies, of which there are various shades, from a very light and 
 almost orange scarlet, to one so deep, as to approach to the crimson. 
 
 The medium between these two extremes, is what we are to work 
 with; for the one tint, by a mixture with orange, approaches to that colour; 
 and the other, by a mixture with violet, becomes a crimson; and though it 
 may be useful hereafter to know how to arrange and manage compositions 
 with these colours, at present we will confine ourselves to the pure scarlet; 
 the harmonizing tint to which is a tint between scarlet and crimson, formed 
 by a mixture of both together, and is a rich and beautiful colour. Should the 
 flower made choice of, have seeds, differing in colour from the petals, of 
 orange or yellow, as is most usual, the introducing a fourth colour is un¬ 
 avoidable; and not to be regretted either, as it will improve and give spirit 
 to the group, for merely red and green alone is not a pleasing mixture. 
 The introducing yellow or orange, gives an opportunity of bringing in either 
 blue, or lilac; and all these colours, if well managed, may produce a very 
 
53 
 
 pleasing effect. The placing the most brilliant colour in the highest light at A 
 and the deepest contrasting tint in the darkest shades at B, should be invanab y 
 done in every composition; and letting the lighter ones take their proper places in 
 the group, where all, if properly arranged, will join in giving effect to the w o e. 
 
 We will now reckon up the tints that may be employed in this group. 
 There are, scarlet in the petals of the poppy, yellow and orange in the seeds, 
 green in the leaves, with blue and lilac in small proportions flung in, as fancy may 
 direct; and these, with all their various shades, are what we have to work wit . 
 
 Suppose A to be a field poppy, so placed that the light shall strike fully 
 unon it and give it all the brilliancy possible ; while another part of the flower, 
 r 1 1 •J in shade should be so much out of the light, as to shew its true 
 W1 , 10U hi § h either t0 o strong a light, or any degree of shade, would prevent 
 
 T In Ind the rest of the flower should be so turned from the light, as 
 being seen, and m R and C I would place more poppies 
 
 " with'that at A ; bat 'have le»s brilliancy: at C there 
 
 :"o.£Tof a weaker tint of scarlet, whose pebats~£ 
 
 case.) incline . IMU «£rose colony,or p,nk, » ^ ^ ^ ^ 
 
 harmonize with, and give value to those of lighter tints. 
 
 , a a.i,lir r pr] to bring in the contrasting tint, by 
 
 It often happens that we a S ' ^ leaf affords us the very 
 
 means of some other flower 01 0 J ’ h gome P Qf a ye il 0 wer shade will be re- 
 
 ^ *** ° f ^ SCarlCt HlaC ’ ^ alS<> 
 as an harmonizing tint to the yellow. 
 
 , a ii the liohter tints should be interspersed 
 
 I need scarcely repeat tha “ h de as the occasion may 
 
 throughout \ and P»«ake of lb. W » »»ad=, 
 
 require, and in the quantities already pointed out. 
 
 J 
 
54- 
 
 APPLICATION OF THE GREEN BLOT, TO A GROUP OF 
 
 FERN. 
 
 In illustrating this theory of colours, it is not necessary to confine myself 
 to flowers, as it is equally applicable to every other object in nature; and 
 there being few flowers whose petals are coloured with green, and of these few 
 none of a favourable tint, the mind naturally directed chooses some other object 
 to exemplify this blot by: but we need not go far in search of one, or even 
 quit the vegetable tribe, for there are numberless plants that, independent of 
 their flowers, are of thenpelves picturesque subjects for the pencil: of these, 
 the common ivy trailing round the decayed trunk or branch of an old tree 
 the wild hop clinging with graceful wreaths and festoons to the larger shrubs 
 for support, together with the whole tribe of graces and ferns, may be enu¬ 
 merated as beautiful subjects ; and capable of being so disposed, as to produce 
 a very pleasing effect; most of them comprising all the tints wanted. 
 
 We will suppose this blot to be a branch or group of fern, growing at the 
 side of an hedge, with some little ground round it. The fern leaf affords all the 
 varieties of green, from the most vivid brilliant colour, next to yellow, down 
 to a very deep shade, next to blue, its harmonizing tint; and to which it may 
 extend, if necessary, to give so much depth to the effect. 
 
 For the contrasting tint, we must apply elsewhere; and its neighbour, 
 the bramble, will afford all that is wanted; for in the leaves of that shrub are 
 to be seen the most beautiful variety of dark reds, sufficient to oppose all the 
 different greens in the fern ; and a small trailing branch of it might, with pro¬ 
 priety, be thrown in, so as to supply all the reds wanted in this composition. 
 The purplish grey, often seen in the heath and around where the fern 
 grows, would very well oppose the almost yellow green in some of its leaves. 
 With these materials, then, we proceed to colour the group : but from the form 
 
55 
 
 of it there arises a seeming difficulty; for the principal of the theory being, 
 that the most brilliant colour will come foremost to the eye, how are we to 
 manage when that colour must, as in this instance, be placed on a retiring , 
 not on 1 prominent part ? for the vivid yellow green in the fern always appears 
 in the young shoots in the very centre of the plant; while the older leaves 
 „rowin« round it, and projecting much forwarder, lose with their growth their 
 first vivid colour, and assume by degrees one of so dingy a cast, as more pio- 
 perly to belong to the second table of colours than the first. 
 
 We must consider, in this first place, that it is by contrast, or opposition, 
 that even white is made either to advance or retire; and it is the same, in a 
 certain decree, with all the other colours : although the power of a brilliant co- 
 attract the eye first, .till remain, to be combated w.th ; and m this, 
 and also in every similar caae, it most be done noth art and judgment 
 
 c P then A to be the situation of the young central shoots of the 
 buppos u green- let the next leaves behind at B, have a 
 
 fern, and of a l ‘ next b eh ind them again broken with a bluish 
 
 bluer and a fuller , ^ Qf tfae game tint> but darker, and lost in 
 
 grey, and those beh ^ contrasting tint be faintly visible m 
 
 a greyish ground, ^ are next to, and lost in the ground, 
 
 those parts of these is ^ next forwarder from A, at C, have 
 
 On the contrary, let os without departing from the natural 
 
 as warm and full ^ gtrong a ’ body 0 f light rest on DDE) (which 
 
 colour of the lea , d jecti leaves ), as their form will admit 
 
 we will suppose to be thre p J * • n _ The parts un der each 
 
 of, and their colour be f^^ontrasting tint flung into the shade beneath 
 leaf, should have as much ground be a mixture of 
 
 it, a. -11 no. h.« XhCld il each oihe,, so .. no, to in- 
 red, grey, and green g &{ bt and colour that should mark 
 
 terrupt the general effe Thc leaves grow i„ g at E E, on each 
 
 the composition at tirst s g blending gradually from the cold 
 
 side, should present a gra ation ^ d on the one side with shade, 
 
 tint at C to the warmer one at B, darkened 
 
56 
 
 and lightened on the other, so as to inclose the central shoot at A, as 
 it were, in a dish or cup. By this means, the vivid yellow green in the 
 centre at A , will remain there , and D D D will come forward to the eye, 
 whilst the other leaves will keep in their respective places. But still the eye 
 will be too much attracted to the centre, and D D D will want more assistance 
 to bring them as forward to the eye, as their natural situations require. 
 For this, we must have recourse to some accidental circumstance, that may 
 with propriety be made use of to assist them: for instance, a young broken 
 shoot of the fern accidentally lying upon them, into which may be thrown 
 a still more vivid yellow green than that at A, which would immediately 
 fling that back and D D D would attract the eye first, especially as the con¬ 
 trasting tints near them would assist and produce the desired effect. 
 
 I mention this, as only one mode of doing it; but the various ways that 
 it might be done, will occur without difficulty to a pencil guided by taste 
 and judgment; for to attempt it by altering or lowering the tone of colour 
 in the centre, and increasing that at B D D, &c. would not be to describe 
 the plant truly, to do which its real colour must be strictly adhered to. 
 
APPLICATION OF THE BLUE BLOT TO A GROUP OF 
 
 BLUE FLOWERS. 
 
 Blue, as has already been observed, is not a favourable colour to form 
 a composition with; but a painter should be able to surmount every diffi¬ 
 culty, as well as avail himself of any advantageous circumstance that may 
 attend the various subjects that happen to employ his pencil. 
 
 To form any thing like a pleasing composition of blue flowers, be 
 must select some of a pale colour, the harmonizing tint to which may be 
 either a deeper blue, or a green, or both; if, as in this case, it be necessary 
 to introduce the latter, which in a group of flowers, cannot be dispensed 
 with, although a composition of blue might be formed without it. Among 
 the variety of blue flowers, there are very few but what have a mixture 
 of other colours in them, although blue may predominate, so as to be the 
 distinguishing colour; and white, yellow, crimson, and lilac, are seen in smaller 
 portions in the petals of all the blue convolvolus tribe. In the minor con- 
 volvolus, the subject, we will suppose, of this blot, both white and yellow 
 form the star in the centre of the flower; and with orange, the contrasting 
 tint to the blue, and black to the white, there will be no less than eight 
 different tints to work with, and distribute throughout the group. Although 
 three of them only are considered as principals, and to be measuied out 
 in prismatic portions, which are the blue, the orange, and the green ; the 
 others, as before directed, are to come in only in the small quantity in 
 which they appear in the flower. 
 
 With these materials, then, I place the most brilliant tint or blue 
 flower at A, in the full light, with some deeper shades of it in more of 
 the flower around it ; the contrasting tint, orange, at B, in the deepest 
 shade ; and it will be easy to select an orange flower of a proper hue, for 
 it should not be of a pure, but of a dingy tint; the green leaves should 
 
 K 
 
be interspersed throughout, but none of them can be of a brilliant green; 
 for, unless of a much lower tone than the blue, they would attract the. 
 eye before it; and therefore care must be taken that neither the green 
 in the leaves, nor the yellow in the flower, are made too striking, lest they 
 interfere and hurt the brilliancy of the blue. The white is friendly to the 
 blue, and will not diminish its lustre; but‘the blue and red must be thrown 
 in very slightly; the latter by means of some accidental circumstance; and the 
 black in as small a quantity as the pure white demands, to assist in the 
 deepest shade, and give spirit to the whole. The reflected tint must be formed 
 by a mixture of the orange, blue, green, and black being thrown into flowers 
 and leaves at the extremity on the shade side, while the lighter tints may 
 be given in the backs or outsides of the leaves and flowers, on all the 
 surrounding parts on the light side, while the brilliancy of the colours are 
 kept in the centre, or most prominent part of the group. 
 
5 § 
 
 APPLICATION OF THE CRIMSON BLOT TO A GROUP OF 
 
 ROSES. 
 
 The great variety of different roses which the gardens now afford, gives 
 file pain teTthe power of exemplifying this blot, either by a pale or a deep 
 shade of the colour, as there is every gradation of it, from the palest pink next 
 to white, down to the deepest shade of crimson, next to violet; and even to 
 violet itself, in a new variety of the flower, named the blue, but more properly 
 the purple rose. 
 
 I shall endeavour to dispose of the whole range, as being a more useful 
 example than a part of it only: but the colour, although beautiful, not being in 
 itself of any great force, especially in the palest degree, the addition of white 
 into the composition, which of course admits black, would be of considerable 
 advantage; as without such aid, it would want that forcetof effect necessary to 
 give it sufficient consequence as a composition, but which the addition of 
 black and white would supply. 
 
 The contrasting tint green, is supplied by the rose leaf, which contains all 
 the various shades that can be wanted. 
 
 The yellow seeds of the rose, which often approach to orange, call for both 
 blue and lilac ; the latter a new variety of the single rose affords, but the blue 
 must be introduced by some other flower. With these materials, or colours, 
 compose as follows. 
 
 At A, a white rose must be placed, so as to receive the light fully; around 
 it there may be others of a mixture of pink and white, with yellow seeds so 
 pale as to harmonize with the white; at B and C some of a full pink with 
 less light upon them; at D there should be a weak mixture of pink, lilac, 
 
yellow, blue, and green, which, if they form a separate mass by themselves, will 
 increase the value of the more brilliant tints in the centre of the group. At 
 E, black and the contrasting tint should be placed; and between them, and the 
 light pinks on the centre, there should be some of a fuller hue, gradating down to 
 violet, in the deepest shade; the reflected tint at F may be composed of grey, 
 and a mixture of the other tints ; but one thing must be observed, viz. that 
 yellow must be sparingly used throughout, and likewise the orange, as neither 
 will do in a much larger proportion in this composition, than as they are seen 
 in the seeds. 
 
6i 
 
 APPLICATION OF THE VIOLET COLOURED BLOT, TO A 
 GROUP OF VIOLET OR PURPLE FLOWERS. 
 
 The different varieties of the common crocus will afford us almost all 
 the shades of violet, or purple, beginning with lilac, the lightest of them, 
 down to a pretty deep shade though not to the deepest; but as we shall 
 form the group of the lightest or palest flowers, a very deep shade of purple 
 will not be wanted for the harmonizing tint; and if we take the wall-flower 
 to convey yellow, the contrasting tint, we may have every shade also, from 
 the palest green yellow, to a full tint, approaching to orange.—The very 
 scanty portion of green in the leaves of the crocus, makes it requisite, to 
 select some flower, whose leaves will afford the necessary quantity, which 
 even the leaf of the wall-flower will hardly supply; and their form, too, being 
 somewhat similar to those of the crocus, a long narrow shape, is a circum¬ 
 stance not in favour of the beauty of them in the group, in point of contour; 
 but those flowers blowing at the same time in the spring, it naturally points 
 out the combination, although with this disadvantage, but which it must be 
 the painter’s endeavour to obviate, by introducing the leaf of some other 
 (flower or plant; and no better presents itself to my recollection at present, 
 than a young climbing twig of the wild honeysuckle whilst in bud, before 
 the flower opens. 'The green leaves like those of the nasturtium, afford 
 two very different tints of green on the in, and outside, which, as they are 
 both of a lighter tint than either the crocus or wall-flower leaf, would serve 
 admirably to distribute the green into the lighter parts of the group; and 
 the honeysuckle, - too, before it is opened in a cluster of purplish, pale 
 yellow buds a sort of broken tint that would conduct the eye to a distance 
 from the principal mass of lilac in the centre and shade side of the group. 
 The unopened buds of the wall-flower also presents a different shade of broken 
 purple tint, as do the stalks, which are often tinged with purple; and all these little 
 circumstances give so many opportunities of distributing the purple or lilac 
 throughout, which it must be more profusely than either the yellow or green; 
 
 L 
 
62 
 
 for these should be eight parts of the lilac or purple colour to four parts five 
 tenths of yellow, and six parts of green. The shortness of the crocus stalks 
 and the delicacy of its texture, forbid the least appearance of pressure in 
 grouping it with other flowers, and would induce me to dispose it in some way 
 that would suit its delicacy, and seem to preserve its freshness, which is injured in 
 a very short time after gathering. I should therefore place them as fresh gathered 
 in a basket, or in an elegant vase: the colour of either of them ought to be 
 a pale tint, of yellowish brown; the palest coloured crocus would then come 
 in at J ; while a trailing branch of the honeysuckle, with its pale yellow and 
 purplish hue in the buds, would intermix with more crocusses of a deeper 
 tbt at B and C, and form the lighter parts; which the wall-flower, either 
 double or single, would, with some of its green leaves, occupy the deepest 
 shade in the group, when of course it would be much subdued in colour, 
 by being in that situation. There are many different degrees of yellow seen 
 in that flower, and one of a favourable tone should be chosen, which should 
 not be one whose petals are marked with the orange brown, but those of a 
 plain yellow hue. The reflected tint at D may be produced by some of the 
 crocusses being flung into shade, intermixed with the wall-flower. I should 
 be very sparing of those of the crocus leaves, for their spiky form would be 
 a disadvantage to the group. 
 
 Although I have endeavoured to make this little work as plain and useful 
 to my readers as possible, I am aware I may still have failed in rendering 
 it as much so to practitioners, as I could wish; for it is scarcely possible to 
 express directions for forming groups, so clearly expressed in writing, but 
 that in some instance or other, obscurity and difficulty will arise to the student; 
 but having however, done my utmost with my pen here, I beg leave to refer 
 the student to my last work, entitled Ornamental Groups, where they will find, 
 in the progress of that work, every one of these imaginary groups exemplified 
 by real ones;* which, I trust will answer the purpose intended, of fully illus¬ 
 trating the theory laid down in this book. 
 
 * The white and crimson blot are both exemplified by groups of flowers, in the first number of 
 the work alluded to. 
 
 F I N I S.