© Site and most content copyright to Peter Weatherill 2017  Some content copyright to other authors as identified
Oil paintings have some protection with the thick layers and the oils they are bound with, but still you will note how many ‘Old Masters’ have reverted from bright colours to browns over the years of being hung in galleries ( Unless they have been lucky enough to be restored ) In many cases the loss of colour is partly dirt and failure of varnishes, but even the thickness of Oil paint is not proof against fading. Acrylics are similarly protected by the thickness of paint and the Polymer binding - and also the modern pigments used in manufacture.  Inks are usually based on more stable pigments, and can be among the most permanent of the water based products but Watercolours are the most fragile of all as the layer of paint on the paper is the most delicate and the thinnest. COLOURED PENCILS In common with other art media, Pencil paintings react with light in much the same way, and need as much care as Watercolours Some pigments are more liable to fade,  and the older reds, purples pinks and some blues are more unstable than the natural browns yellows and greens.  With modern chemistry and the organic colours produced from the oil industry, we now have a range of colours that are more stable in strong light.   For this reason, many old stocks of pencils contain a high proportion of unstable pigments, and newly produced stock tends to have better stability as pencil manufacturers are now more conscious of the need for colours that will last. Take Care, though.  Most manufacturers produce a wide range of pencil lines and not all will be made using the more expensive light resisting pigments.  School and student lines will often be made with cheaper ingredients, and low price ‘own brand’ coloured pencils sold by stores who do not specialise in art materials, may well be more subject to fading in many colours.   Those manufacturers who produce lines specially made for Artists and meet high lightfast requirements often have the highest priced products, and this is for a reason - the pigments are more expensive. Manufacturers also like to market a full range of colours even if some colours in the sets have low lightfastness. Not every buyer of coloured pencils is either aware - or even interested - in whether the results of their work is liable to fade. Speaking to the manufacturers, I find the common factor is that they make the highest profits on the lower cost items sold for schools and children.  The highest quality art materials made for artists are often those sold with the smallest margin. They are manufactured and sold to gain the quality name that sells the other stock. WHY DO WATERCOLOUR PENCILS OF THE SAME COLOUR AND BRAND AS WAX PENCILS, CARRY DIFFERENT RATINGS ? A pigment used in a wax based (non soluble) pencil will usually have a higher lightfastness rating than the same pigment in an Aquarelle (Watercolour Pencil). In an aquarelle, the eventual thickness of pigment once it is brushed in with water will be less than when it simply laid on the surface with a wax binder.   The pigment is the same (usually), it is just that the thinner layer of colour in the aquarelle is more delicate and liable to fade. In July 2013 - Caran d’Ache introduced a new softer Aquarell into the UK.   Museum Aquarelles are soft, strongly pigmented and also lightfast. They have been sampled and tested and results on the handling are included on the Caran d’Ache page.  The pencils have a high standard of lightfastness both dry and wet, and therefore have very good durability with ratings of 3 stars and above on the Blue Wool scale. ARE THERE COLOURS TO AVOID ? Because the amount of pigment in pale colours is lower than the quantity in stronger shades (due to the mixing of more filler with the pigments),  paler colours - particularly blues and pinks, tend to have the lowest protection against fading. Many older brands of pencil have lower ratings. but some newly formulated pencils can have low light ratings too. IF YOU WANT YOUR ARTWORK TO LAST IN DISPLAY, YOU NEED TO CHECK THE STANDARD OF THE COLOURS YOU USE      Refer to the manufacturers websites and literature.   You will usually find some reference to the lightfastness (UV rating) where you find the colours listed    If a ‘Blue Wool Scale’ rating is shown, the highest rating is 8 - the most lightfast -   and anything of 6 and above is considered  acceptably lightfast.   Personally I try to avoid colours rated 4 and below and consider 5 as acceptable. Some pencils carry a star rating and I try to avoid any with a 1 star mark and use the 2 and 3 star pencils.   Try to check out the rating either from the pencils ( Faber Castell and Caran d’Ache ) or from the manufacturers website ( Derwent ).  If you can’t get hold of a rating, treat pinks, purples and reds with caution - especially the pale colours which, as noted above,  very often have low pigment levels.    Strong light can encourage the little pigment that is there to fade. Some brands conform to ASTMS 6901 and carry reference to LF1 and LF2 gradings.  This is tested on the actual pencils and as such is the highest test rating.  LF 1 and 2 are excellent - one is higher than 2, though !   See further down the page for more information on ASTMS 6901. One or two brands show testing to ASTMS 4303. This is testing on the actual colourants used in the pencils and whilst it is not a test of the pencils, it is a good indication that the colours you lay down will last. YOU DON’T WANT TO SPEND MANY HOURS ON YOUR ARTWORK TO SEE IT ALL LOST AS THE COLOUR DISAPPEARS LINKS TO MANUFACTURERS WEB PAGES GIVING LIGHTFASTNESS SETTINGS FOR THEIR PRODUCTS      NOTE :  You will need to copy the Internet address where given below, removing the initial brand title, and then paste it into your internet browser I am indebted to a correspondent in Slovenia who has taken some time in researching web sources for pencil lightfast ratings  The list he has provided is included in the revised list below ( March 2011) I have not had the opportunity to check these links for the new Topics site, so cannot guarantee they all still work ! DERWENT  ( Use the blue wool scale - see above ) Derwent Coloursoft: http://www.pencils.co.uk/files/productchart/Coloursoft%20Colour%20Chart__j5zjn4i0.pdf Derwent Inktense: http://www.pencils.co.uk/files/productchart/Inktense%20Colour%20Chart__nzrfmgu0.pdf Derwent Artists: http://www.pencils.co.uk/files/productchart/Artists%20Colour%20Chart__p1dk1gz5.pdf Derwent Studio: http://www.pencils.co.uk/files/productchart/Studio%20Colour%20Chart__sjfv2kvn.pdf Derwent Drawing: http://www.pencils.co.uk/files/productchart/Drawing%20Colour%20Chart__ftzxjveb.pdf Derwent Metallic: http://www.pencils.co.uk/files/productchart/Metallic%20Colour%20Chart__m5kwe50u.pdf Derwent Graphitint: http://www.pencils.co.uk/files/productchart/Graphitint%20Colour%20Chart__qaxffxe0.pdf Derwent Watercolour: http://www.pencils.co.uk/files/productchart/Watercolour%20Colour%20Chart__zynx3axe.pdf Derwent Aquatone: http://www.pencils.co.uk/files/productchart/Aquatone%20Colour%20Chart__frzmhc5c.pdf FABER CASTELL   ( use a star system - see above ) Their definition of their lightfastness ratings is at Faber Castell star system Link     ( the original Faber Castell site link has disappeared - July 2014) Details of the pencil colours is to be found at Faber-Castell Polychromos: http://artgraphic.fabercastell.com/ASSETS/C63C35093C0D4FAE97814B277666921D/ColorChart201 0.pdf Faber-Castell Albrecht Durer: http://artgraphic.fabercastell.com/ASSETS/C63C35093C0D4FAE97814B277666921D/ColorChart201 0.pdf CARAN D’ACHE  ( also use a star system for brands other than Luminance and Museum) The information on lightfastness of the different brands is available by downloading PDF files from these links : Caran d'Ache Pablos: http://www.carandache.ch/m/la-couleur/artistes-et-professionnels/les- crayons/pablo/index.lbl Caran d'Ache Luminance 6901: http://www.carandache.ch/m/la-couleur/artistes-et- professionnels/les-crayons/luminance-6901/index.lbl Caran d'Ache Supracolor: http://www.carandache.ch/m/la-couleur/artistes-et-professionnels/les- crayons/supracolor/index.lbl Caran d'Ache Prismalo: http://www.carandache.ch/m/la-couleur/artistes-et-professionnels/les- crayons/prismalo-angle-artistes-/index.lbl Like Faber Castell, use anything with a single star with caution. Luminance ratings are given as LF1 and LF2  - according the the ASTMS 6901 data.   LF1 is 100% lightfast and LF2 is 80% lightfast  - I would regard all Luminance colours as fine for CP work. PRISMACOLOR Sanford have been very coy in the past about publishing the data. However  they have now published an official Prismacolor Premier Chart showing the colour card and the official ASTMS 6901 ratings for their main 132 colours ( and blender ) They show the range with the 5 grades listed from excellent ( 1 ) to poor ( 5 ) On this scale, the serious CP artist would normally be aiming to use those in grades 1 and 2 and there are 72 colours match this aim, with quite a few of the ‘good’ colours, shades of grey or white. For the PDF file of the Official ASTMS Prismacolor chart see: http://www.prismacolor.com/Style%20Library/PrismaColor/media/promotions/Prismacolor_Lightfa st_Color_Chart.pdf OTHER BRANDS Staedtler and Lyra do not publish data - or if they do, I haven’t found it in any leaflets, but  UV star ratings for Lyra are shown against each colour inside the lid of the box. TALENS ( VAN GOGH ) Talens Van Gogh pencils meet the standards of ASTMS 6901 though not to the level of Caran d’Ache Luminance, there is a chart of Van Gogh colours and ratings ( by stars ) here :   Van Gogh pencils: http://www.talens.com/english/downloads/colour_charts/VGCP_GBR.pdf There are also technical papers on lightfastness, ASTMS and Blue Wool standards here : http://www.talens.com/english/technical/default.asp?subID=1&subsubID=1.1&subsubsubID=1.1.3 &subsubsubsubID= And a chart showing the conversion of Van Gogh star ratings to ASTMS here: http://www.talens.com/english/technical/default.asp?subID=1&subsubID=1.1&subsubsubID=1.1.3 &subsubsubsubID= From which you will see that the Van Gogh 2 star verges into LF 3 I also understand the Dutch made Bruynzeel FullColor pencils are lightfast, but I have no data and they are not available in the UK CRETACOLOUR Made in Austria by the successors to the Hardmuth pencil business. Colours of the 36 Karmina ( dry point ) and Marino ( watercolour ) pencils meet the ASTMS 4303 standard with all colours either LF1 or LF2 Karmina can be very difficult to find, but Marino are on sale in the UK and I have tested them as very good ( and not highly priced ) ASTMS ratings for lightfastness ASTMS 6901 relates specifically to coloured pencils and is product based   i.e testing is done on the actual pencils  ASTMS 4303 relates to lightfastness of pigments and artists colours in general and relates to the ingredients. ASTMS 4303 Testing is not done on the product, it is done on the ingredients ASTMS 6901  -  ASTMS 4303 These notes quote from AMIEN.Org web site http://www.amien.org/forums/showthread.php?1578-Summary-of-ASTM-Standards ASTMS 6901 ASTM D 6901, “Standard Specification for Artists’ Colored Pencils” This standard is a departure from ASTM D 4302, and applies only to colored pencils. The difference here is that artists’ colored pencils are mixed-colorant products. Therefore, these materials require product-based testing. That is, the actual manufactured product (a colored pencil) is tested by the manufacturer of the colored pencil. This is very different from the methods of ASTM D 4303 that involve testing the coloring materials used in the products. The principles of ASTM D 4303 apply, but the sample preparation and test methods are slightly different. Of course, therefore there are performance and property requirements for these materials that make them different from those covered by D 4302. This standard is a significant departure from ASTM D01.57’s traditional testing scheme, and therefore will be used in the development and writing of other, future, standards that require product-based testing ASTMS 4303 ASTM D 4303, “Standard Test Methods for Lightfastness of Colorants Used in Artists’ Coloring Materials” This is the technical method that describes the ways in which colorants (pigments, dyes, lakes, and so on) used in most single-colorant artists’ paints can be tested for their relative lightfastness. It describes two types of light exposures used in the method, how to prepare samples for testing, the conditions of the exposures, the instruments used for measuring the light intensity during exposures, the instruments that can be used to accelerate exposures, the specifics of the languages used to characterize color and the instruments that can measure color using the color languages specified, how to measure the colors, and how to interpret the results. Two theories can be simply explained here. 1.) Many artists use at least some white paint in their mixtures – or paint thinly on white surfaces. This is an assumption based on observation of artists at work in their studios and the observation of works of art in display collections. ASTM D01.57 therefore specified that single-colorant paints would be made with the colorant mixed with a standard white for testing purposes. That is, ASTM D 4303 specifies the physical nature of the materials being tested. 2.) The second theory is that accelerated testing is correlated with the passage of real time – the test method reflects a reasonably accurate picture of what would happen if 20 – 100 years of testing would occur. The theory was first tested in the 1970s, by Robert L. Feller at the National Gallery of Art, in Washington DC, and reported in a paper published in 1978. To date, the theory has been verified in a number of unpublished studies. A Research Report on the test methods can be obtained from www.astm.org: request RR D01-1036. Coloring materials tested with this method receive a “Lightfastness Rating” of I, II, III, IV, or V, depending on the results. Only Lightfastness Ratings of I and II are good enough for artists’ materials. It should by now be clear that this kind of reporting of results is not predictive – ASTM D01.57 does not predict the “service life” of materials.
PLEASE BEAR IN MIND that any links to other web sites quoted here are only good for as long as the web site at the other end of the link remains unchanged.  Web sites are updated and page contents change.  Some information moves elsewhere on a site and the link then shows up as failing.  If you spot a link that is no longer valid and can trace the new address for the same information, please let me know.
FINALLY If you wish to preserve the quality of the colour in a Coloured Pencil Picture, you can take further steps to protect the image. You can spray the finished work with a fixative that also adds a protective Ultra Violet light filter and will therefore add a layer of protection.   You can also have the picture mounted and framed up behind a special protective glass or plastic which will also provide more UV protection. Both these protections will add a filter between the picture surface and the viewer,  so some colour will be affected
BEFORE LOOKING AT PENCILS, LET ME FIRST LOOK AT OTHER MEDIA AND DISCUSS HOW THEY REACT WITH LIGHT The thinner the layer of colour on the surface, the more liable is the pigment to fade or change in bright sunlight.   For this reason completed watercolours are carefully stored out of bright light and displayed with caution.  
In June 2018, a comprehensive set of tests have been published here which originate from several years of work by Judith Crown. Sets of sample strips from most brands have been subjected to the strong sunlight of Israel for upwards of 12 weeks at a time - equivalent to many years of more temperate UK sunlight.  The resulting tests are shown on the adjoining lightfasteness page.  They make good reading, but please read the forward to that page carefully
LINK To the CROWN Tests