We are asked to use the same subject/composition for this and the next exercise that we used for the colour accuracy exercise. For this one, we are to create a painting using only two colours (complementaries) and maybe the addition of white.
I drew an accurate representation of the source image, using a trace off technique from the photo, which had been printed onto A4 paper. I lightly shaded the reverse of the photo with a 6B pencil and then traced off the main lines onto Bockingford paper using a 4H pencil. I did this because I had a feeling I might end up doing several version of the same image in different colour combinations and didn’t want to have to spend too much time on the drawing part of it.

I initially felt a blue-violet/yellow-orange pair would work well and this time.

I found it quite tricky to get the tone up (or down) to the right levels and in some places went over with a white wash to dull it down. I wasnt entirely convinced with the result. And in retrospect, I realised that I had purposefully not mixed the complementary colours to create a grey, which is maybe where I should have started from!

I decided to have another go with red/green complementaries- using water-soluble oil pastels, again on Bockingford paper (which I find is one of the best to use for a lot of wet mediums like watercolour and water-soluble oil pastels, ‘cos it can take the working about on its surface without degrading). I still tackled the painting the same way – i.e. using different huses of green and red to build up the tones.

I didn’t like the result at all. It was lifeless and indistinct. I then decided to mix green and red to see what kind of a ‘grey’ I’d get. It was a lovely burnt ochre, so I watered it down a lot and went over all the green areas, in various degrees of pigment strength.

I have done a lot of smaller experimental tone and colour studies of this arrangement, in between making the paintings, so I’m starting to get a little bit sick of it!

When I look at all the studies now (as they are on the wall (see next exercise), I am finding that the red/green study seems to work the best – but they all have a uniqueness.