Painting Roses and Colour Mixing: Sheffield Florilegium Workshop

I recently ran a workshop on painting roses for The Florilegium Society at Sheffield Botanical Garden. It's always interesting to talk to participants and take on board what they have to say about where they find difficulties. I offered to write this blog post for those who attended on the day and for general interest for anyone interested in drawing and painting roses, with a few additions on palettes and colour mixing too which was a common topic of the day.

A rose painted before the workshop, colours worked out in advance using a combination of Permanent Rose, Quinacridone magenta and a touch of cobalt blue in the shadow colour. The use of blue in a shadow on a pink rose gives a very fresh look, which is important in roses and ther flowers.

It was a very warm welcome from quite a large group of 18, I probably talked far too much but hopefully shared some useful information.
Drawing and painting a rose in a day is a tall order at the best of times, so I had forwarded some preparatory information of how to approach the drawing of a rose, but as you no doubt know, there are an enormous variety of roses, however the basic principles for drawing and painting are the same.

How to approach a Rose Painting
If I were to paint a rose for a florilegium, I would probably paint it over two seasons. This is the best way to familiarise yourself with the plant. I take the approach of a study page initially, which enables me to 'get to know' all of the plant parts as well as working out the colours, approach and watercolour techniques. For the workshop participants were to draw and paint just one aspect of the rose but hopefully they will now be better equip to tackle roses in the future. Here's a breakdown of the process I outlined on the day.

A study page started at the Sheffield workshop. The rose is William Lobb, a Moss rose

Line Drawing
If the drawing is flawed or incomplete, it certainly wont get any better as the painting progresses and this is particularly true with a complex form, such as a rose. Consequently an incomplete drawing can easily become confusing and rather overwhelming.
The best approach is to measure and be systematic, break down the subject into shapesand don't be afraid of taking photographs for reference... remember roses like to rearrange their petals when you're not looking! I always start by taking overall dimensions, then look for noticeable angles, which I plot, next I look for 'land marks', such as unusual shaped or distinctive petals. I gradually plot the petals on a basic grid, which allows me to navigate around the flower and to see the petals in relationship to each other by position. I work on tracing paper for ease of corrections transfer to watercolour paper this mehtod aviods wasting paper so doesn't cost much if you make a mistake.

Stages of Drawing: Measure, plot angles using pencil and thumb method and find 'land-mark' petals to work your way around the flower in a systematic way.

Palette and Colour Mixes
Start with finding the basic hues for all parts. Having a piece of white card with a hole cut in it can help you to isolate the colours in the subject, remember that surrounding colours of an object can alter your perception of colour, so it's important to remove other colours that may distract you. You can take photographs for reference but never try to match colours from a photograph - it will not be accurate, instead always colour match from life in good natural daylight by a window.
Identifying colour mixes seemed to be a general problem for many people on the day. My attitude is to keep it simple, work with a palette of 12-18 paints. A good range of single pigment primary colour paints is what should be used for florilegium work, remember the whole point of this type of work is longevity and documentation, and that means using the best materials available. Paints should have ASTM I or II,  Opera Rose should never be used for Florilegium work! a few participants did use it for fun though and by way of comparison. Some people say they have done windowsill tests but a test of 5 - 10 years has no robustness, so best to be cautious and avoid fugitive colours....if the manufacturer say its fugitive, I'd take their word for it. I start with colour and technique experimentation on small areas, such as petals.

A simple exercise in working out colour and technique, for William Lobb, a David Austin Rose. Start with a simple petal on your study page or sketchbook. Colours used here: Quinacridone Magenta, Permanent Rose and Violet Dioxazine. Three washes and dry brush work to create the texture of the petal. 

Greens
Greens always seem to present problems, I have seen some of the most bizarre colour mixes, with multiple pigment paints and even using colours with black, such as Indigo and Sepia, this is bound to end in a flat and dirty appearance, so again keep it simple.
Painting a green chart is useful (see below) and mix all greens from primaries.  I don't recommend using convenience colours such as Sap Green because almost every different brand has different combination of pigments, some have up to 4 others have 2 and none in common with each other, the result is unpredictable! Every time you add another pigment to it you move nearer to mud! I'm not saying never use convenience colours but if you do use it make sure you add no more than one additional pigment.
Carefully observe the colour to see how it is affected by light, this can make it warmer or cooler and there may also be underlying colours.

Careful observation of greens is required! I would use and underlying cobalt blue wash for the highlights in this leaf. Some areas are more yellow biased ( leaf stalk and mid rib) and some more blue biased. See how the light catches each leaflet, making one side of the leaf blade darker and more blue than the other blade because of the effect of light.

 It's much better to mix your own greens I start with a 1:1 ratio of a blue and a yellow to mix a secondary colour and usually add a small amount of red to make a more natural green. Then I can bias the mix from cool, with more blue; to warm, with more yellow, and also towards a more muted or appearance, with more red in the mix. Here's a short clip of how I work out greens ( not specific to the rose but hopefully still useful. I can also mix all the browns and any other colours this way and with the same three colours. 

I usually mix greens using a blue with a light value similar to the green that I'm trying to achieve, no-one at the workshop had ever heard of this before! so we discussed it at the end of the day. Basically, this means that I use a high light value blue, such as cerulean for a light coloured leaf, a mid light value blue, such as cobalt for a mid value green and a dark blue, such as Indanthrene, for a dark green leaf. Below you can see the high light value greens on the top row, the mid range in the middle row and the the darks on the bottom row. Of course you can paint them in different densities to make darker of lighter too.

Green mixes separated by light value of the blue. Top row: High light value blue mixes, middle row, mid range blue mixes and bottom row, dark values blue mixes. This simple process allows you to identify the type of green that you are looking to achieve with your leaf.

 Always make a note of the colours and paint a swatch for future reference and as part of your working practice.

Here's my suggested palette: approx 5 Reds, 5 Blues and 4 Yellows plus a few others:

• Reds: Permanent Rose, Permanent Alizarin Crimson or Permanent Carmine, Quinacridone Magenta, Permanent Magenta and and Scarlet lake

• Blues: Cerulean blue, Cobalt blue, French Ultramarine, Winsor Blue and Indanthrene Blue I also like Prussian Blue
• Yellows: Lemon Yellow Nickel Titanate, Winsor Lemon, Winsor Yellow and Transparent Yellow
• Others: Cobalt Violet and Violet Dioxazine

Don't overwhelm yourself by buying too many paints, it's important to understand the properties of the paint and to learn colour mixing first.  You can add more later but you don't really need them. 

The Painting Preparation and Process
Paper: Always use 100% cotton archival paper. Transfer the drawing by tracing lightly on the back of the final drawing with a B grade pencil. Use a soft tip fine liner to trace on the front side to transfer and this will allow you to see where you have been with the tracing and will prevent you from applying too much pressure, which can indent the paper and ruin it.

Brushes: Don't use brushes that are too small, this tend to lead to an overworked appearance and makes very hard work of it. I use a Kolinsky sable Size 5 round and pointed/ round for the initial washes and size 3, 2 and 1, miniatures or spotters for detail and dry brush (I use old worn brushes for dry brush work), also series 107 Pro Arte spotters are great for detail. I also find the Pro Arte synthetic Series 62 Flat shader, size 1 excellent for repairing edges and lifting.  Again I have other brushes but this selection will do the job These last two brushes are cheap craft brushes and cost about £1.75. There is no need to purchase specialist brushes for this purpose. Most are simple cheap brushes that have been re-branded.

Use appropriate sized brushes, I've seen artists using size 2 brushes for washes, this is far too small

 Watercolour Techniques
I use a range of techniques but generally employ a staged approach as follows:
1. Apply the 'tea' wash sometimes wet-in-wet or wet on dry in smaller areas, this lays the foundations. Be sure not to have pencil line under the wash and remove as much as possible first. you can also work just inside any pencil lines on the outside edge.Then remove any remaining pencil after the first tea wash.
2. Build colour with more layers of washes and introducing some detail and shadows, more selected washes in the areas requiring more depth of colour. washes
3. Add detail and dry brush work to build texture and final detail
4. Tidy up any areas such as edges if needed.

On a complex rose I find it easier to work from the outer petals and work towards the centre, as shown in the video below, I also work on alternate petals so that they are dry before moving to the next petal.  

Using a burnisher is useful between washes, as with some papers 'flocking' occurs on the paper surface. Make sure that the washes are completely dry before using. I use either a flat agate palm stone or my new toy which is a gilders burnisher, which is excellent.

A gilders agate burnisher is perfect for smoothing and flocking of the paper surface, which can happen with certain papers. Available from gold leaf suppliers

Overall it was a very enjoyable day at Sheffield. I learned about the the Society and you can find out more and view the archive of work on their website

30 Paintings on Vellum in 30 Days

Haven't written a blog post for over almost a month and it's really bothering me! There's one in the offing about photographing artwork but it requires some use of the grey matter to finish it off, so here's s something about the new 30 Day Challenge in the meantime. It was initiated on the Botanical Artist's Facebook Group, and requires 30 small works to be completed during September. Over the past few years I've tried a few of these Challenges, here's a link to one that I actually completed in 2013, and wrote a daily blog for.
This time I wanted to try something different so have decided to do all 30 on vellum....it is do-able if I keep it simple!

Day 5-10 Violas on Kelmscott Vellum, size approx 8 x 6. I bought these little pots of flowers from the local DIY store and painted the whole sheet in one day because I was so behind. The palette included: Holbein Manganese Blue PB33 ( not the hue) with W & N Cobalt Violet and Quin Magenta. I love the way PB33 separates, it's really useful for high light value pale violet flowers, with both pink and blue  present. Also Violet Dioxazine. Lemon Yellow NT and Cadmium Yellow for the flower centres. The greens included Cobalt Violet, Lemon Yellow and permanent rose. 

 The Challenge started while I was away in Italy, where I visited the Venice Biennale, it's good to view a diverse range of art and the Biennale was certainly that! diversity definitely feeds the brain and upon my return I was was left playing 'catch-up' with the Challenge but definitely inspired to work.

I try to tie in this sort of challenge with other work, and these little studies are prefect for video examples for my new online Vellum Course which is underway with 5 great new students. Currently I'm also working on a large vellum painting of Roses on vellum so no doubt a few roses will probably feature here and there. Here's a little bud of Olivia Rose Austin, which was my 4th painting.

No 4. Olivia Rose Austin Bud. 5 x 3 inches on Kelmscott. Colours: Permanent Rose and Lemon Yellow n.t. for the pink with Cobalt Violet for the cool pink shades. A small amount of Permanent Carmine was also used to deepen the colour . The green mix is Cobalt Blue, Transparent Yellow and Permanent Rose. The stem Perm rose, Trans Yellow and Brown Madder

 The Blackberry studies are good as preparatory works for another painting, which is for a  competition, the final piece will be on a piece of mounted natural vellum though. This is more challenging than the Kelmscott but this piece is all good practice with the subject.

Day 1. Blackberry Study on Kelmscott, approx 7 x 5.5 inches. Love painting blackberries and working with dark colours. The black mix is Indanthrene Blue, Permanent Carmine and a very small amount of Trans Yellow. The bias of the mix is on the violet side. Where the fruit is more of a dark red colour, the quantity of Permanent  Carmine is increased. The red is Perm Carmine and Scarlet lake for the warmer red. Stems and all other parts are mixed from the same 4 colours. There is really mo need for lots of colours when a few will do and the use of the same colours created better harmony in a painting.

No. 3 This is a simple study of a single fruit and leaf but painted with Ivory Black. This is a very thick piece of vellum 6 x 2 inches. I quite like this effect and have been playing with it recently. It's all about getting the tone right before working with colour. It's also an approach that I teach as good practice in preparatory work.

Also completed on the first day was a simple single maple seed, not particularly well painted but very quickly completed.  I used the earth colours for this, it was a pretty lazy effort. In the 2013 Challenge a maple seed was the first subject and it seemed a good idea to repeat on vellum.

Maple seed on Kelmscott, 3 x 2.7  inches. I used some Manganese Blue on the shine. Thereafter Raw Umber wash, followed by building texture with Brown Madder and Van Dyke Brown, a small amount of Ultramarine Violet was added for the shadows. Lots of sweeping and drawing dry brush work here.

I've almost caught up! Maybe I'll even manage all 30. Ive plenty of vellum off - cuts so no excuses.

The Photographing artwork post will hopefully follow shortly.

Everything's Coming up Roses! The First Studies

A few weeks ago I visited the David Austin Garden, which is less than an hours drive from my home.  In the past I haven't been a huge fan of roses and only ever illustrated a few in all the years I've been painting.... but decided it was time to give the roses another chance. The garden is probably one of the best I've ever seen and got me completely hooked on roses. I purchased three plants and the painting commenced! Roses seem to be one of the more challenging subjects in the world of botanical painting, so I'm open minded about the success of this project. Will split this into two separate posts, this one is about my preparatory studies for Olivia Rose Austin and Munstead Wood.

First study page of Olivia Rose Austin

I had intended to buy an old rose but got drawn in by 'Olivia Rose Austin', which is an English Leander hybrid and has only been available since 2014. With its large pale pink cupped blooms, this rose is named after David Austin's granddaughter, so I guess they must think it's pretty special too! It blooms three times a year instead of the usual two, so I'm hoping that buys me some time. This rose along with a couple more,  Munstead Wood and Sceptre d' Isle, should keep me busy for a while. 

Olivia Rose Austin in full bloom in late June at the David Austin Garden in Albrighton

I couldn't write a post without a few photographs of the David Austin Garden, so here it is.....a feast of roses!

The Renaissance Garden

Beautiful archways and climbers can be seen throughout the garden. There are several different areas, including a Victorian walled garden, here the doorway looks in from the Long garden to the Lion Garden. 

I took well over 500 photographs and several videos on my iPhone. I returned again the following week for another look and although it was just one week later it's wasn't quite as spectacular. There is also a species garden and it was interesting to note how many bees were found in this area. Here's a short clip of typical of a very busy little bee. If the current paintings go well ( fingers crossed!) I'd like to paint some of the Scottish species roses next year.

Where to start: the study page approach for Olivia Rose Austin
I plan to paint the final works on kelmscott vellum and have a whole skin waiting on top of the wardrobe. But my initial approach is to complete sketchbook and then a number of study pages, the study page is the focus of this post. These studies are painted on A2 Fabriano Artistico watercolour paper ( I40lb HP) but a study page should be painted to suit the size and structure of a plant. 

I don't plan a layout for a study page but place the elements randomly with fairly even spacing on the sheet.

Because the final piece will be on vellum a number of small colour studies on off-cuts of vellum will be completed too, I do this because the colour will be different on vellum compared to paper. This study page approach allows me to get to grips with the subject and stimulates ideas for a composition. It's not painted in great detail and probably takes 2 or 3 days to complete. But I don't set a limit...it takes however long it takes.  Most importantly though it enables me to get to know the subject and to understand the character of a plant, such as the growth habit, and in this case the changing morphological features and colours of the blooms.
Painting study pages is particularly important to me especially for a plant that I'm less familiar with. It's similar how a portrait artist needs to capture the character of a sitter, so the botanical artist must capture the particular character of a plant. I use this approach with all of my own work and all of my students are 'forced' to do it too. Those student who really commit to study page work, instead of skipping over it,  seem to get the best results in the long term.

Detail from page 1 of the study: Different stages of flowering, with faded older blooms, spent flower and bud.

Colour test on vellum. I paint these directly to vellum 6 x 4 inch off-cuts without drawing, they take about 30 mins to complete. I paint the flower rather than just swatches of colour because the layering of colour needs to be understood in an actual subject. I make notes on the sheet and keep for reference, noted include the colours and descriptive information, such as the fact that the opening bud is a stronger pink than the open flower. Fortunately not much adjustment in colour was needed for vellum, slightly less yellow was required because of the base colour of the surface. 

 In a study page the aim is to collect as much information as possible, written descriptive notes and illustrations of the plant parts that may be useful, such as different aspects of the flower, buds, leaves etc. sometimes I make several pages of studies, basically I continue until I'm happy that I've enough information - there isn't a set formula, it's just common sense. This enables me to work through many of the problem areas before committing to the final painting and reduces the number of potential errors. It's an approach that's probably not suited to everybody - I think mabye having a scientific background reinforces my own methodical approach to planning a painting. Having said that I don't always do a study page and sometimes I complete spontaneous paintings but with new subjects for major works, it's a wise move. 

 How to tackle a complex form
A rose of this type has a complex structure of many petals, after making some initial sketches I developed a reasonable understanding of the petal arrangement. I decided after putting in a feint outline in, that it was easier to sketch directly with watercolour. Then working from the outside to the center seemed to work ok, but keeping careful track. The lack of pencil helped me to keep it very clean and avoided any erasing. I only ever draw in very feint outlines use an eraser as little as possible - even light use can damage the paper surface. The washes need to be very light and carefully layered in a pale flower and detail is added with dry brush.

Beginnings. A few pencil guide marks and then paint lightly. I used W & N Permanent Rose mixed with Lemon Yellow nickel titanate to give a soft pink. The yellow is opaque but this works to the advantage with a soft pink giving a creamy soft pink. There is a lot of discussion about not using opaques amongst botanical artists and while I agree that most useful colours are transparent, the opaques also have important place in the paintbox, their value is dependent on the order of washes and the light value of the subject ( but that's another post). I wanted to capture the softness of a rose and so this seemed like a good option. I also added Permanent Magenta for cooler pink and used M.Graham Cobalt Violet for the shade. (note: that despite being the same pigment, the M. Graham CV is much cooler than the W & N version).

Gradually work into the centre building up the colour

A confusing aspect of rose paintings is the change in the flower, they expand from a fairly simple bud and end up with a mass of petals. The colour changes quite dramatically too, so I painted the rose at different stages of maturity.  

The leaves are quite a challenge with roses and of course any study needs a number of leaf studies. I have a few simple ones on this sheet but will complete more before proceeding with the painting. 

A simple study of a leaflet. The leaves are all important to any painting, so many paintings are ruined by poor leaves. Roses are pinnately compound leaves, here just the terminal leaflet is shown.  I think rose leaves are possibly one of the most difficult to get right. There is generally lots of fore-shortening involved too, with fine veins and some puckering in the older leaves! I always tackle the leaves first on the painting because to me if the leaves aren't good enough the painting is ruined. For these I used a mix of Indanthrene Blue, Transparent Yellow and Permanent Rose. For greens I pretty much always try to use transparent colours. I keep the mix to 3 or less colours of single pigment and generally use a small amount of 'red'  to make the green more natural colour. If the red is one I used in the flower I use the same in the green mix.

I also did some research on other paintings of roses, from the obvious Redoute paintings to some contemporary painters. One of my favourite paintings of a rose is Martin Allen's beautiful single pink rose, also Kate Nessler's roses on vellum are old favourites of mine.


Munstead Wood 
I wanted to capture all of the roses and they were all flowering together, so decided to complete the study pages for all first. Munstead wood is an English old rose hybrid and very different rose from Olivia with its rich dark wine and magenta coloured blooms. It's another beautiful fragrant rose and reminds me of childhood days spent collecting rose petals from my grandmothers garden.  

Munstead wood study page underway. A very dark flower which changes colour from a crimson bud to a deep magenta or burgundy flower.

A difficult coloured flower, that changes colour over time. I used a combination of Permanent Magenta, Permanent carmine, Violet Dioxazine. Also for the bright glazes Quin Magenta and Quin Rose. An a little neutral tint for the deep shade.

The open bloom has many petals and appears two-tone in colour as it changes.

I've a little more work to do on the study pages and will then move on to the composition, so will continue with this post at a later date to describe the process of composition. Thereafter the final painting on vellum will begin. Hopefully I can complete it this year while the roses are still flowering. If not I prefer to wait until the following year when the plants flower again. I could continue from photographs but only do this if I absolutely have to. 

One thing I have noticed is that painting roses is a messy business! All of a sudden the petals drop on mass! they fall into the water and on to the palette, resulting in this rather nice effect!

Petals in the paint, leave beautiful patterns