Plant tolerance of dry heat. Part 1.

Patrick Mills / This article was first published inThe Mediterranean Garden, the journal of the Mediterranean Garden Society, Nº 39, January 2005.

Years ago, glancing through a nursery catalogue in a friend’s house, I happened upon a section on old roses and then the words, “some do well in poor soil”. Ah, I thought, this could be for me (my soil is ghastly) and immediately wrote off for ninety-six roses. For the moment I had forgotten that soil was only one quarter of my problem: the other three were – and still are – severe cold, extreme heat and no water. The cold didn’t bother the roses much but the other two factors played hell with them, of course. So, in order to turn a threat into an opportunity, I rationalised that what I was really doing was carrying out a valuable experiment to find out which roses would stand such a hostile environment. And I proceeded to plant a couple of hundred more.

In order to describe the roses’ resistance to this environmental threat, I needed to characterise their behaviour in summer. Two categories of behaviour were immediately obvious: those that died and those that looked awful. But this was not very helpful. I needed to be able to differentiate more among the survivors, to be able to see which ones seemed to suffer less, which ones looked better (or less bad) in a garden in severe Mediterranean conditions. I had to have a classification system of some kind. But before going more deeply into that, let me tell you a little story.

“Once upon a time there was a farmer who had a lazy son. One day he called his son and said, ‘I am going to town today and I have an important job for you. You see this big pile of apples here? I want you to sort them into three piles. Put the big ones on this side of the barn and make a pile of small ones on the other side. Put the middle-sized ones in the middle here. Oh, and any bad ones you find, throw them in that corner over there. Make sure you get each apple in its right pile. When I come home tonight, I’ll look at the piles and if I see any apple in the wrong pile there’ll be no dinner for you tonight.’ So the farmer went off and the son, grumbling, started sorting the apples, examining each one and then putting it in one pile or another. As he had to look carefully, he occasionally found one that was starting to go bad and threw it into the corner; and more than occasionally he retrieved an apple from one pile and put it in another. By mid-afternoon he had finished but, as he loved his dinner much more than he hated work, he spent the rest of the afternoon going backwards and forwards from one pile to another, looking very, very carefully until he was absolutely sure that not a single apple was out of place. At sunset his father came home, glanced at the piles and then got some sacks. He proceeded to fill each sack, putting in a few big apples, a few middle-sized ones and a few small ones until the sack was full. At this his son got very angry: ‘Why did you make me do all that work for nothing?’ ‘It wasn’t for nothing,’ the father replied. ‘The important thing was to separate the bad apples. But if I had just asked you to pick out the bad ones, you would not have looked properly. You would have missed half of them. That is why I made you look very carefully at each and every one, by making you decide exactly which pile it belonged to.’”

So what is the relevance here? For the moment, just forget about the bad apples; they are not the point. For us the point is that being obliged to decide on which category something belongs to makes us examine it much more carefully – and the more categories there are, the more carefully we have to compare and contrast it with the rest. The farmer’s son could have been given five categories or even a dozen. If I wanted a classification system for describing my surviving roses, I would first have to examine them to establish a yardstick by which one is judged superior to another, then create a number of “piles” (grades, classes or categories) and finally examine each rose carefully and contrast it with its fellow to decide on its grade, constantly cross-checking and revising the grades.

But why just roses? Why not all garden plants? A greater variety of plants would manifest greater differences and it would therefore be easier for me to start with them and learn to classify or grade their behaviours. So it was that I set out to find a system for defining a plant’s resistance to our Mediterranean summer conditions. As this is a subject central to our garden planning and practice, I present it here for everyone’s consideration and interest. First there are a few thoughts on what we consider acceptable plant performance and what we do not welcome; next, a reminder of what the stressful elements of the Mediterranean summer really are; then some comments on existing basic systems of classification and their limitations. The main body of the article describes the grading system I have been using and its applications, together with a sample list of a hundred plants contrasting their summer behaviours according to my observations so far. Part 2, to be published later, begins with a description of an alternative system (devised by the American Horticultural Society) and its problems. There is some further comment on the difficulties of the system I use and some concluding remarks on the objectives of the article.


Measuring performance: The "dry-heat resistant coefficient"
When I look into my garden right now (July), I see a row of five cistuses: the first on the left is totally dead; the second is a brilliant, bright green; the third is three-quarters dead, literally; the fourth is not too bad but has lots of leaves turning yellow, ready to drop; and the fifth has lots of dead twigs – about one third of its parts. Regardless of whether this is caused by heat and drought or not (probably not, in this case), which plant would you prefer in your garden? I know that some people quite like dead bodies in their gardens (“It’s part of the natural process”) but I am afraid I don’t. Lots of annuals are dead (naturally) by midsummer and we usually clear them away; the leaves of spring bulbs die down and we try to hide them with other plants. It is true that a few plants leave remains – whole or partial – that are arguably quite as interesting as the living thing: the statuesque spike of Ferula communis, for example. But a plant dying of an overload of stress is part of an unnatural process: being placed in an inappropriate environment by a human gardener. (Or is that “inhuman”?)

Most of us prefer not to have the garden full of dead and dying plants, so my guess is you will prefer cistus no. 2 above, and all such similar plants, the same as I obviously do (although somehow I do seem to end up with lots of the other kinds too). At the height of the hottest summer, in the midst of dead leaves, dead grass, dead garden plants and an overall depressing brownness, anything that is a bright cheerful green – even without the help of irrigation – is a true boost to the spirits, a sign of life and optimism. Why don’t I have more of these things in the garden and fewer of the gasping, trying-to-die, can’t-stand-this-heat type? There are two common reasons why this might happen to us: firstly, ignorance – we lack information on a plant’s capabilities and behaviour; and secondly, wilfulness – we plant in spite of the advice we are given. We’ll never do much to correct the second situation but here I hope to address the first. How can we know more surely which plants can best withstand the conditions of our Mediterranean summers and talk more easily about them?

To start with, what are these conditions? What is different about our summers? Basically, it is hot and it doesn’t rain (much). So this is what plants have to put up with:

Dry air: low humidity (some places and
times)
Drying air: hot winds (more stressful than we usually realise)
Direct radiated heat: the pitiless sun
Reflected heat: from stone walls, buildings etc.
Dry soil: lack of rain

Please note that there are two basic factors involved, heat and dryness together. We are not talking about tropical jungle heat; nor are we talking about cold Siberian drought. The plant has to cope with both high temperatures and no rain (or very little), so that we must talk about a resistance to dry heat and it is this dry-heat-resistance of the plant that we must be able to describe and grade.

“But don’t we already have such grading systems?” Well, sort of. We can, for example, say a plant is “drought-resistant” or “nearly drought-resistant”. This is a bit like saying “cold-resistant”. The question is, whose cold, mine or yours? 0C, minus 5C, minus 15C or what? “Drought-resistant”, similarly, does not tell us a great deal. A better example is the system used in some gardening books, of putting one, two or three shaded water drops beside the plant’s name to show watering needs, or an empty drop to show it needs no watering. Very neat indeed for publishing purposes but not so good for other types of communication: what do we say, a “one-drop” plant, a “three-dropped plant”?

Much better still are those simple systems using a scale with few categories and defining them in simple words, such as “Very drought resistant”, “Tolerant of drought”, “Appreciates a little water” and “Moderate watering necessary”. These are much more useful and almost good enough, for many purposes anyway. But I have two problems with them. First, I think we can go a little further than this, or even much further if necessary. And second, perhaps more important, I find it perverse to describe the plant’s own inherent resistance in terms of watering in the garden, as if irrigation were the solution to everything and the amount of water the only yardstick. This is precisely what we in the MGS are supposed to be challenging. The plant has a degree of resistance, which we should be able to gauge and describe, and once we know it, we can decide whether to plant it or not, and if so where to plant it – what sun it will get, how much shade and when, what winds – and lastly, what water, if any, it will receive.

For my own part, I chose to represent resistance by giving each plant a number, using the scale of zero to five, with the possibility of using all the decimals in between if necessary. This is what I call the plant’s “Coefficient of Dry-Heat Resistance” and I think of it like this:
“A measure of the degree of resistance of a garden plant to prolonged exposure to dry heat as demonstrated by the absence or presence of negative effects – wilting, shriveling or discoloration of leaves, defoliation, dieback, etc – regarded as unwelcome or unattractive in the garden.”

This “coefficient” or “index” is a constant, for practical purposes: we expect two strong and healthy examples of a certain species or cultivar to behave in the same way. If we plant a pair of, say, Lavandula angustifolia ‘Twickel Purple’ side by side, we expect one to react just like the other but perhaps differently from other species in the same site, or even differently from other cultivars of the same species. What we need to do is to grade plants’ resistance in relation to others: which ones are more resistant and which ones are less, like the different-sized apples, but instead of three piles of apples, we have as many categories as we want to use. We can talk of a plant having a resistance of roughly “3” or “4” or “5”; or we can slice more finely and talk about “2.9” or “4.1” or “4.8”. We can say that a group of plants – the genus Lavandula, for example – are “around 4”; or we can compare them, rank them and find that different cultivars may be graded as 3.7, 4.0 or 4.4. Such differences in performance are quite noticeable and worth knowing about, to take into account in future planning.

So how do we establish these relative resistances in the first place? We must first observe a good number of different garden plants growing under the same conditions more or less typical of the Mediterranean, compare and contrast the reaction of different plant species or cultivars and rate them, like the apples. In the year I started to create my “league table” of dry-heat resistant plants, these were the conditions which the observed plants had to face in July and August (as in most years in my garden):
• Daily maximum temperatures normally 30-35, occasionally 25-30 or 35-40. Very occasionally over 40C
• Rainfall of 15-20mm in 4-5 showers
• NO SUPPLEMENTARY WATER
The list on pages 47 and 48 will show you a hundred or so plants together with the “coefficient” or grade I think each plant warrants. These grades are in absolutely no way to be considered definitive: they merely represent the story so far. Some plants were observed for one season only, some for two; most were in one site only, but some in more than one. Like the farmer’s son, I must go back and check again and again. The plants are not sorted: trees, shrubs and other garden plants are all lumped together. After the main list there are a few native and agricultural plants for comparison.

The first question one would ask is, “What do all these numbers mean?” The simple answer is that a coefficient of 4.5 is more dry-heat resistant than one of 4.0, and much, much more resistant than one of 3.0. But people are hoping for a more concrete answer: what does it mean “in real terms”? And the answer is not so straightforward because growing conditions vary and so “reality” – what we see out there in the garden – also varies. Let’s take this slowly, step by step.
First, how might plants react to summer conditions like those outlined above? Here are some descriptions of possible reactions arranged on a scale, from ‘better’ to ‘worse’:
• No visible effects
• Deterioration more or less acceptable
• Significant unwelcome effects
• All parts above ground die off: plant possibly dead
• Lasts a few weeks but eventually dies if not helped
• Wilts and dies within a week or so
• Can’t survive these conditions under any circumstances.

The desire is naturally to link these reactions directly to the dry-heat coefficient numbers but unfortunately there can never be a one-to-one universal relation for the simple reason that (now please… READ…THIS…SLOWLY):
while a garden plant’s inherent drought resistance is for practical purposes constant, its reactions can vary when growing conditions vary.
This change in growing conditions can be natural, a milder summer for example, or not natural, a gardener watering the plant or giving it shade. I can tell you what the numbers mean in most of my garden most of the time (and I will); I can also give you a fair idea of what they might mean in your garden. But a number can’t mean exactly the same reaction everywhere all the time as the plant obviously will react differently to different conditions. If we go through a few of the numbers, I think this will be clearer.

Let’s start at the top. The ranking of “5.0” I reserve for those plants that could stand up to conditions far worse than those mentioned, perhaps even those in Death Valley, California, for instance; totally invulnerable, for practical purposes. A little further down the list at “4.5”, plants such as some rosemaries sail through those mentioned conditions (no watering, remember) without blinking an eye: “no visible effects” in my case and probably in most of the Mediterranean too – but not in Death Valley! The plants in the list on page 47 that say “4.5+” are those that I strongly suspect could withstand conditions somewhat worse than I describe – in an exceptionally difficult site, for example – and still show no ill effects. Those that I have marked “4.6” or “4.7” have already performed in such an exceptional site without a murmur – a performance well beyond the call of duty, you might say.

Those plants rated round the “4.0” mark do show the effects of stress, more noticeably or less, but such that one can tolerate them. Teucrium fruticans (4.0), for example, one could never imagine to be in danger of succumbing to heat and drought anywhere in the Mediterranean but it does change its appearance and begin to look a little the worse for wear. But it can be easily forgiven.

Plants around “3.5” may cause me to worry: Cistus x canescens (3.5), for example, “Is it going to get worse?” Or make me, unreasonably, annoyed: typical santolina (3.4) for example, “Why are you looking so bedraggled and boring? You are supposed to be a native around here”. Plants at such grades may look happier in your garden if your natural conditions are better – or if you give them goodies.

It is in the middle of the range, say between 2.5 and 3.5, that variations in growing conditions, brought about either by nature or by human hand, can have the most noticeable and perhaps crucial effects. In a normal summer in my garden, “3.0” means the line between life and death (the “deadline”?): the plant will die if it gets no help from the gardener. Campanula muralis (C. portenschlagiana) (2.9), for example, only survives here because I give it shade and a little dose of water if it looks to be on its last legs. However, in an unusually benign summer, as 2004 seems to be so far, those plants graded as “3.0”, or just a little below, may very well survive, though showing evidence of considerable stress. And if your natural conditions are less severe than mine, perhaps your normal “deadline” will be more like 2.7 or even 2.5, and not 3.0 as in my case.

Just as some plants in this middle range may be saved by more favourable conditions, they can be endangered by a harshening of conditions, again either by natural or human intervention. In most of my garden, Erigeron (3.2) will normally just get by without help but it may not survive a particularly bad summer or – and this is important – it may not survive a normal summer if I put it in a particularly difficult site in the garden. (See “microclimates” in Part 2 of this article.)

Those were some examples of how plants of different coefficients of resistance might behave, depending on the conditions. All those ratings you see in the list (3.1, 3.2, 3.3 etc) I established with the “apple system”. At the end of August I go around with my notebook, assessing damage and grading plants or checking previous grading. To evaluate a plant’s performance I examine it carefully, make a comparison with other plants and argue with myself: “This plant B looks like a 3.7. No, hold on, Plant A is a 3.7 and it does look a bit better than this. So perhaps B is a 3.6… or maybe A should be 3.8. Let’s check with a few others in the 3.6-3.8 area.” And so on. It definitely makes me sympathise with the farmer’s son (it is easier if you have someone else to argue with), but I am gradually accumulating information.

You will have noticed there are no plants in the list in the bottom range of numbers, although I was supposed to be using a scale of 0 to 5. I could have planted some typical northern European garden plants and then observed them die more slowly or more quickly, but I didn’t. Someone else can do that; I did enough of it years ago. For the moment, let it be sufficient to say that a plant with a very low coefficient will eventually die in the open garden, even with its roots constantly moist. Unless you build an air-conditioned room round it.

The next question is, “What is the use of all this? Especially if my conditions are different from yours.” Just suppose for the moment that this list of plants and their dry-heat resistance coefficients were tried and tested, reliable (which it isn’t) and ten times longer. Here we would have solid information we could use when placing plants in the garden; or, putting it the other way round, information to use when selecting plants for a particular site. No scale is going to tell you exactly how a plant will behave in your garden: you yourself still have to know your own garden, its problems and possibilities. Once again, to put it crudely, “Resistance is constant but conditions vary, so reactions vary”. A gardener in a somewhat more user-friendly situation than mine – because of cooling breezes, morning mists, shade from trees or topographical features, underground streams or whatever – might say something like this, “I find that, generally speaking, in my garden a plant with a coefficient of 2.6 or so will probably survive without watering, although they will look pretty scruffy. This little chappie here, though, is only a 2.2. I can’t do without him, that’s why he is here by the kitchen door where I can keep an eye on him and give him a drink when he is in trouble. But up there, on that slope among the rocks, I would not risk anything below 3.5, not in that heat. Better still, 3.8 or 4.0. And down at the gate, at the beginning of the drive, I have to put 4s or 4.5s. They have to look good all the time and there is no emergency service down there.” This gardener knows his or her garden well and makes good use of information. A focus on the plant itself and its inherent resistance draws attention to planning and selection, placement and protection and only in the last resort irrigation. It’s a question of getting the right plant in the right place rather than post hoc attempts to correct “mistakes” (lack of information), impulses or whims.

The interesting thing – and the main reason for writing this article – is that, if we in the MGS could agree in the first instance on the grades of only 50-100 plants commonly used in Mediterranean gardens, using the system described here or any other comprehensive system, these could then be used as “criterion plants” or “yardstick plants”, so that further plants could easily be slotted in by simple comparison of performance with criterion plants in the same site and the same season, whatever the reigning conditions happened to be. Over time, information on hundreds of plants could be built up, a priceless store of information. It is difficult to imagine a more useful or practical project for the MGS as an institution to sponsor.

However, no system is, or ever will be, perfect and the second part of this article, to be published in the next issue, will examine some of the difficulties that arise and objections that may be made. (So don’t write in just yet!) .

Pistacia terebinthus: 4.7
Jasminum nudiflorum: 4.7
Cupressus macrocarpa: 4.6
Sedum spectabile: 4.6
Atriplex canescens: 4.5+
Cupressus sempervirens: 4.5+
Delosperma cooperi: 4.5+
Perovskia atriplicifolia: 4.5+
Rosmarinus officinalis
‘Boule’: 4.5+
Teucrium chamaedrys: 4.5+
Cedrus deodara: 4.5
Coronilla glauca ‘Citrina’: 4.5
Euphorbia characias: 4.5
Euphorbia pithyusa: 4.5
Origanum vulgare: 4.5
Rosmarinus officinalis
‘Corsican Blue': 4.5
Rosmarinus officinalis
‘Majorca Pink’: 4.5
Rosmarinus officinalis
repens: 4.5
Stipa tenuissima: 4.5
Teucrium cossonii: 4.5
Viburnum tinus: 4.5
Atriplex halimus: 4.4
Cupressus arizonica: 4.4
Broussonetia papyrifera: 4.4
Dorycnium hirsutum
‘Fréjorgues’: 4.4
Koelreuteria paniculata: 4.4
Helichrysum italicum: 4.4
Lavandula x intermedia
‘Dutch’: 4.4
Lavandula angustifolia
‘Twickel Purple’: 4.4
Pistacia lentiscus: 4.4
Rosmarinus officinalis
‘Tuscan Blue’: 4.4
Rosmarinus officinalis
‘Baie d’Audierne’: 4.4
Sophora japonica: 4.4
Teucrium divaricatum: 4.4
Teucrium flavum: 4.4
Thuja orientalis : 4.4
Ziziphus jujuba: 4.4
Ceratostigma
plumbaginoides: 4.3
x Cupressocyparis
leylandii: 4.3
Lavandula x intermedia
‘Alba’: 4.2
Rosmarinus officinalis
‘Punta di Canelle’: 4.2
Teucrium marum: 4.2
Santolina chamaecyparissus
‘Pretty Carol’: 4.2
Artemisia vallesiaca: 4.1
Lavandula x heterophylla
‘Devantville': 4.1
Lavandula x
‘Richard Gray’: 4.1
Artemisia arborescens
‘Powis Castle’: 4.0
Artemisia canescens: 4.0
Dorycnium pentaphyllum: 4.0
Euonymus japonicus: 4.0
Landula angustifolia
‘Seal’: 4.0
Lavandula angustifolia
‘Grappenhall’: 4.0
Phormium tenax: 4.0
Salvia microphylla: 4.0
Teucrium fruticans: 4.0
Thuja orientalis ‘Aurea’: 4.0
Artemisia arborescens: 3.9
Phlomis purpurea ssp.
almeriensis: 3.9
Ballota pseudodictamnus: 3.8
Phlomis fruticosa: 3.8
Lavandula dentata: 3.7
Tanacetum densum: 3.7
Artemisia lanata: 3.6
Cistus albidus: 3.6
Cistus x aguilarii: 3.6
Thymus capitatus: 3.6
Ceanothus ‘Autumnal
Blue’: 3.5
Erysimum ‘Bowles
Mauve’: 3.5
Cistus x canescens: 3.5
Populus alba: 3.5
Nepeta faassenii: 3.5
Senecio viravira: 3.5
Teucrium creticum: 3.5
Artemisia ludoviciana
‘Valerie Finnis’: 3.4
Santolina
chamaecyparissus: 3.4
Thymus carnosus: 3.4
Achillea filipendulina: 3.3
Centranthus ruber: 3.3
Medicago arborea: 3.3
Ruta graveolens: 3.3
Syringa vulgaris: 3.3
Cistus x florentinus: 3.2
Dorycnium pentaphyllum
ssp. herbaceum: 3.2
Erigeron karvinskianus: 3.2
Salvia forskaohlei: 3.2
Salvia fruticosa: 3.2
Salvia officinalis: 3.2
Acanthus mollis: 3.1
Brachyglottis ‘Sunshine’: 3.1
Caryopteris x
clandonensis: 3.1
Clematis cirrhosa: 3.1
Stachys lanata: 3.1
Teucrium hyrcanicum: 3.1
Artemisia frigida: 3.0
Iberis sempervirens: 3.0
Plantago cynops: 3.0
Rosmarinus officinalis
‘Barcelona’: 3.0
Salvia argentea: 3.0
Campanula muralis: 3.0
Gaura lindheimeri: 2.9



And for comparison:
Kermes oak: 4.9
Olive: 4.8
Holm oak: 4.8
Phillyrea: 4.8
Hackberry: 4.4
Grape: 4.3-4.5
Local elm: 3.8
Dog rose: 3.4
Almond: 3.4