Odd colour mixes that work ...

For one, the Zorn palette - Yellow Ochre, Vermillion or Cadmium Red (I strongly suggest the second of those) Ivory Black, and white.  It shouldn't work, you might think, but in oil it does; it even works in acrylic, which doesn't always behave in any way similar to oil paint.  I've yet to try it in watercolour - I'm not at all sure that black in w/c is going to work; but I can't see any theoretical reason why it shouldn't.  A mix I tried today, in watercolour, was recommended by Jeremy Ford, in a booklet I've recently bought from the Michael Wilcox School of Colour - for flesh, in one of his paintings, he used a mix of Pthalo Green, and Cadmium Red Light: I tried it today, and - it works!  I don't know if the painting will come together, I've allowed it and myself a rest overnight, but I see no reason bar my own incompetence why it shouldn't.   Do you have paint mixes that instinct tells you won't work, but in actual fact do?  I tend to be a bit conservative in colour-mixing, so would welcome anything you lot out there can come up with as worth a try.  The science of it all is suggestive - but perhaps most of us rely more on instinct and past practice than science. The Michael Wilcox School recommends, and sells, a maximum of 12 colours as being entirely adequate to achieve virtually any colour or tone you wish to achieve: they break down the colours into rational categories - e.g. violet-inclining blue; orange-inclining yellow - and it does work, very well; though I admit to introducing other colours from my dissolute past as an enthusiast for traditional hues.  It does save money though - and it does work - if you stick with the Wilcox formula, of White; Burnt Sienna; Quinacridone Violet/Permanent Rose; Cadmium Red Light; Yellow Ochre; Raw Sienna; Cadmium Yellow Light/Pale; Hansa Lemon Yellow; Pthalo Green; Cerulean Blue; Ultramarine; Pthalo Blue.  Not a hint of black - but with those Pthalo colours that our wonderful Syd Edward hated with such passion. Now - I know I've posted something similar to this before: and I'm still not one of Michael Wilcox's shareholders.  I do not restrict myself to this palette; I'm fonder of Viridian than of Pthalo Green (both are great mixers: faintly hideous on their own), I would substitute Cobalt Blue for Cerulean (usually), and while I think Pthalo Blue is excellent with watercolours, I prefer Prussian Blue with oils - and marginally, with acrylic.  But the principles are the same, when you break them down to green-inclined blue, orange inclined yellow, etc.  What profiteth it me to recommend this approach?  Well - I've seen SO many student artists who think they need to buy dozens of colours: and this is fiendishly expensive in oil paint - that every now and then, in an uncharacteristic fit of altruism, I feel bound to suggest to beginners AND more experienced painters that you just don't need the 120-plus hues available in the Old Holland range, or the slightly less extensive  Winsor and Newton and Talens Rembrandt offerings.  In fact, you need - as in really NEED - just 10% (if my maths is right: it rarely is) of that.  Think just how much you could save on your next paint purchase if you bore that in mind?   If you can mix your own violet; your own orange; your own greens (very easy with those colours), rather than buy them ready-made in the tube, you could save tens of pounds, and over a year of painting, hundreds of pounds.  I know we're all artists, with our heads in the clouds, but that's got to be worth doing, hasn't it?   My paint box contains many more colours than this - a quick look shows me, in oil, Indian Red, Indian Yellow, Light Red, Venetian Red, Cobalt Violet, Mars Violet, Mars Brown, Mars Red, Mars Orange, Chrome Yellow Hue, Cadmium Lemon, Terre Verte, Ivory Black, Rose Madder, Brown Madder Alizarin, Permanent Mauve, Winsor Red, Rowney Golden Yellow, Sap Green - and that's a QUICK look - there are certainly more.  And we haven't even touched on the variety of whites I use.  A lot of those colours aren't necessary: they can be mixed, and mixed quite easily at that.   Choose the colours you want - as indeed I do.   But don't bankrupt yourself if you don't have to: fine if you're as rich as Croesus, but most of us aren't; a limited palette like this will give you greater unity, AND save you maybe a thousand pounds a year, once you tot it all up.  In gratitude, you can buy Dawn Farley a birthday present with all the money you've saved!

A wooden palette is a favourite of mine, for oil at least: I have found the disposable ones too slippery, yes.  As for mixing - proportions are important: eg, I'm not sure I'd want to use that pthalo green plus cadmium red for flesh again: it's a bit on the brown and pale side, and gets muddy if you need to strengthen it; but again - the secret will lie in the proportions and thus your (Norrette's) plan to practise on odd bits of paper is probably essential: it's a hard lesson to learn if you mix up the desired colour, apply it, and find it doesn't work on an actual painting.  Normally, I'd use burnt sienna, raw sienna, touch of cadmium red, plus whatever I need for shadows, which could be anything from Indian red to Viridian, to Ultramarine.   Syvia, you are perennially youthful!  What's more, you have developed an eye for colour after much experience: the fact that you can tell an orange red from a bluish one - not everyone can -is testament to the fact that you've moved beyond the need for formulae and systems.   Oh, and on the subject of red - pyrrole red is marketed as Winsor red by W & N: it's a very good red to have, and I use it quite a bit: cadmium red is more orange: I love it!  A discreet splash of either in a painting is a great way to bring it all to life.   Quinacridone violet - a lovely colour, a less blue cast than Alizarin Crimson, and much more lightfast - comes to us under a number of guises: Permanent Rose; Rose Madder Hue; Quinacridone Red; Quinacridone Magenta -  the identifier we want is PV.19.  (Last time I looked, the SAA sold a Rose Madder Hue, I don't think they called it quite that, with a different pigment - which I remember not liking very much: I'd stick to PV.19.)

It think it's more the grinding of the pigment than the medium it's ground in, given the amount of medium doesn't vary by much: but I don't make paint, so take that as a very inexpert opinion.  PR108 IS a pure pigment -  perhaps I've misunderstood what you mean (very likely I've misunderstood what you mean!  I'm good at that...).  There are cadmium red 'hue' colours available -  none of them, irrespective of what their makers say, can match up to genuine cadmium red.  Though they have their uses - rather few, in my book.   It's fairly easy to tell orange- and blue-leaning reds apart - colours like Vermilion, Cad Red, Cad Scarlet, Scarlet Lake, tend towards orange (which means they carry an element of yellow, basically) while crimsons tend to blue: which would include Alizarin Crimson, Crimson Lake, Permanent Rose/Quinacridone Red, Violet, Magenta, and a number of others.  Pyrrole red/Winsor red has a useful tendency to swing both ways .... if you'll forgive the comparison...   It's a rich red of considerable subtlety whose properties others can probably explain better than I can - well worth having in the palette, because it darkens very well, can be enhanced with yellow, to make a rather splendid orange, and modified with blue or green to make violets in the first case and darks with both.   The snag with not knowing which reds will do what arises particularly when it comes to mixing - eg, if you want a bright orange and don't happen to have a tube handy: you'll get very different results from different reds and yellows - it saves a lot of time and paint if you know what will give you the desired result first time.   I recommend the Handprint website for a lot of information on colour - it deals with watercolours, but the chemical composition of the paint is what we should be looking for, and the reliability of the base pigment in each case.  I like the site because it doesn't faff about: it tells you very clearly which pigments to avoid, as well as those which should be your first choice.  

Cara - interesting point you raise there, eg Quinacridone orange, 'what's the point?'.  There are a good many colours that are packaged for convenience which could easily be made by elementary colour mixing, Quin orange being one, Cadmium orange another.  I'd defend Quinacridone gold, though (not that it's been attacked): it would be very hard to get near to a colour mix equalling it.  It's instructive to take a close look at paint labels - if you see your tube of Indigo is actually a mix of, say, Pthalo Blue and Black - and you've got both in your regular palette, there's not much point in buying tube Indigo: and the real thing is fugitive anyway.   Dioxazine Purple - I agree with you entirely!  Cobalt Green - the painter David Curtis had an article in The Artist a good few years ago in which he extolled its virtues as one of the only greens you could use out of the tube to represent foliage - I've never used it, the price is a bit steep, but would like to give it a go one of these days.   A colour I can well do without is Cadmium Green - a) you can mix it, b) would you really want to?  Though it'll vary between makers.  Don't care for Cobalt Violet either - which is just as well given its cost: another colour much better mixed in general - though I've seen it very well employed in sunset/sunrise paintings.