Variation under Domestication

This semester I am taking a course on Darwin’s first edition of On the Origin of Species.  I thought that I would post chapter summaries of the Origin as I read it, to help keep the material straight in my mind and hopefully introduce you to one of the most important texts of the past 150 years.  If you are a creationist, perhaps you will find less in this book to fear than you had thought; if you are a scientist, you might be surprised by how prescient Darwin was in many ways.  If you are neither, you will still be impressed with the power of Darwin’s arguments.  If you would like, you can follow along by reading the Origin with me – the first edition is free online at http://embryology.med.unsw.edu.au/pdf/Origin_of_Species.pdf. 

I will be skipping over the opening quotes (read my thoughts on them here) and the introduction, and will dive right into the meat of the thing.  This one's a long one...it should be shorter in the future!

Chapter 1 – Variation under Domestication

Some examples of wild species (left) and their domestic varieties (right).  From Konrad Lorenz (1973)

Darwin was not exactly what you would call the most entertaining writer in the world.  He’s certainly not bad – in fact, when he gets excited about something his writing becomes almost poetic, and there are numerous quotes that elevate my imagination and set my heart racing.  But then there is the beginning of the first chapter, arguably the most important component of any book since you risk boring the reader right off the bat, and unfortunately that is exactly what Darwin does.  Rather than begin with his shocking assertion that species evolved from other species, rather than say as Dawkins does in the first sentence of The Selfish Gene, ‘I have now made it possible for there to be intellectually-fulfilled atheists’, Darwin starts with a whimper.  But it is an important whimper, as he lures his reader in with one problem and eventually reveals that within this problem lies the key to speciation.

Darwin begins by talking about farm animals.

Particularly, he says that individual farm animals tend to be more diverse or varied in form than individual wild animals.  So farmed salmon, for instance, display more variety than wild salmon.  This is interesting, since wild salmon were thought to be a production of God, while farmed salmon were a production of man.  Although Darwin doesn’t put it in these terms, the question he is really asking is, how does man produce more variation than God?

The answer, Darwin suggests, has to do with climate and food.  Wild Atlantic salmon, for instance, are found in some rivers in a particular part of the world.  But farmed salmon are found all over the world – something about these different climates may reveal or produce new variations.  Wild animals must struggle for food, while farmed animals are well fed – this, too, may somehow create new variations not seen in nature.  Today we would likely say that the farmed environment, being climactically different from the wild, and having low predation with an abundance of food, both relaxes selection (allowing mutations that would never make it in nature to survive) and alters the selective regime (such that variants not favoured in the wild are favoured in the farm), while exposing recessive mutations.  With my farmed salmon, I saw this all the time – bent spines, double tails, strange yolk sacs would not prevent those individuals from surviving, even though they would never last in the wild.  Darwin, however, was unable to explain how these two factors (climate and food) produced greater variation among domestic than wild animals.  The real answer is that variants survive in the farmed more readily than in the wild environment, not that the farmed environment produces more variation. 

This is one of the frustrating things about reading Darwin – he wrote before the genetic revolution.  He spends a lot of time talking about the source of variation, and is often so close and so wrong at the same time.  It’s not his fault, how could we blame someone for not knowing what nobody knew?  Darwin had no means of knowing how new varieties could arise, but today we know about mutations in the DNA, and how these mutations can lead to new types of variants.  What Darwin did notice was that farmers never seemed to exhaust the possibilities for their domestic productions.  Even though wheat has been domesticated for thousands of years, new forms of wheat are still being made today.

A lack of knowledge did not prevent Darwin from speculating.  Right away, in only the second paragraph of the book, he tries to provide at least some clarification regarding the production of variation.  He blames variation on changes in the reproductive system of the domesticated parent, prior to the offspring being conceived.  In this he was actually somewhat correct – many of the processes responsible for variation occur during the production of sperm or egg cells.  But he suggested that somehow the very act of domestication altered the reproductive system, thereby making the parent more susceptible to producing variants.  As proof, Darwin noted how domestic plants were often nearly sterile, or how rarely animals kept in captivity bred, even if they were still in their homeland.  Something about domestication seemed to affect reproduction, and for Darwin this, however it worked, produced varieties.  ‘Sterility has been said to be the bane of horticulture; but on this view we owe variability to the same cause which produces sterility; and variability is the source of all the choicest productions of the garden.’  Darwin was wrong, but it was a nice try.

There was an observation working against Darwin.  Some forms of sports (new suddenly arising variants of plants) came from grafts, rather than from the parental plant’s union of pollen and ovum.  If changes to the reproductive system caused variation, then how could new varieties arise out of grafts?  Simple, says Darwin – buds and ova are physiologically similar.  Thus the point is proven – new varieties are formed through changes to the reproductive organs or features similar to reproductive organs, and must occur prior to fertilization (since grafts are formed without fertilization).  Although this entire conversation is misguided, it does reveal Darwin’s incredible ability to take observations that seemingly counter his arguments, and through a simple logical flourish turn it into proof of his argument.

So, variation occurs quite commonly, argues Darwin, among domestic organisms.  But, interestingly, he argues that the role of the environment is actually quite small in generating this variation.  Although plasticity may play a role, especially among plants, in producing different forms (that is, a certain fertilizer may cause wheat to grow differently than if that fertilizer had not been used), it seems that some greater principle of inheritance is at work to produce these varieties.  A single berry may produce numerous seeds, and each of those seeds may be grown under environmental conditions that are identical to that of the parent; even so, they will not look identical.  Something unknown is causing them to vary, and to vary irregardless of the external environment.  How much Darwin would have benefited from the writings of Mendel (the father of genetics)!  How DNA would have enhanced Darwin’s argument!

What laws regulate this variation?  First, says Darwin, use and disuse (Lamarckian evolution) – domestic ducks tend to have heavier leg bones and lighter flight bones than do wild ducks, supposedly because the domestic ducks walk more and fly less.  Domestic cows have enlarged udders for milking, presumably because they have been constantly milked.  Domestic animals have droopy ears, because they no longer need to remain alert to danger.  Use and disuse may cause individuals to vary, and this could in turn be passed on to their offspring.  Second, variation is regulated by a mystery known as ‘correlation of growth’.  That is, if one part varies, other parts may vary with it.  Cats with blue eyes also tend to be deaf; creatures with long limb tend to also have long heads.  Something, somehow, links certain traits together such that a change in one by necessity changes the other.  This is today known as pleiotropy, and occurs when the product of a single gene has multiple effects (or, alternatively, linkage disequilibrium, when two genes are located so closely together that their two separate products are rarely if ever separated during meiosis).

‘The result of the various, quite unknown, or dimly seen laws of variation is infinitely complex and diversified.’  Although Darwin cannot fathom how variation is produced, it exists, and it exists in copious amounts.  ‘Any variation which is not inherited is unimportant for us’ – thus if a bodybuilder works hard to be strong, this will be of no consequence to his child, who will have the muscles of a regular person and not of a bodybuilder.

Most variation, however, argues Darwin, is heritable.  Imagine that a father and a son are compared – they will likely share a number of features.  However, did the son inherit these features from the father, or are they shared because the same environment acts on both the father and son?  Says Darwin, we can never be sure.  However, if the father has a rare condition found in one out of every million people, and the son also has it, the likelihood that this condition was inherited increases dramatically.  This is even more true if the father and son are in different environments.  Since albino dads tend to have albino kids, it would seem that albinism is heritable – and if albinism is heritable, than much of the variation we see is likely also heritable.  Variation, therefore, is copious and is passed down through the generations.

The laws governing inheritance, like the laws governing variation, were unknown during Darwin’s time, but they could be glimpsed through observation.  Some had argued that inheritance was regulated through blending – the mom’s traits and the dad’s traits blend together in the child.  Darwin, however, argued that blending could not be the dominant rule for inheritance, since reversions are quite common.  A reversion occurs when a child has a trait found in the grandparents but not in the parents.  How could this happen if child is a blend of the parental traits?  Reversion is therefore an important aspect of inheritance.

Even more important is the observation that a peculiarity found in the parent will also be found in the offspring, but not until the offspring has reached the age at which the parent first displayed the peculiarity.  Thus if the parent acquired a disorder when she was fifty, the child will not display the disorder until the late forties or early fifties, if the disorder is heritable.  Once again, these observations skirt around genetics without there being a proper vocabulary in which to discuss them, but seem to show that traits are handed down in discrete packets of some sort, rather than being simply blended.

After this brief foray into inheritance, Darwin returns to domestic organisms.  A common belief at the time was that domestic organisms, if released into the wild, would exhibit reversion – they would revert back to their wild state, proving that these domestic ‘varieties’ had really changed very little from their ancestral stock.  A domestic chicken would look more like a fowl after several generations of being in the wild; a German shepherd would look more like a gray wolf.  But, argues Darwin, how does anyone know this to be true?  Most domestic varieties would not survive in the wild (imagine a flightless duck, for instance, surviving in fox-filled territory for very long) – those that would survive would have to be those most like the ancestral stock to begin with!  Furthermore, for most domestic animals the ancestral form isn’t even known.  For instance, is a Chihuahua descended from a wolf, or a fox?  It is therefore quite difficult to know if reversion has occurred.  But if reversion does occur, argues Darwin, it will not be because the variant is going back to its original state; rather, selection in the wild is operating to favour individuals more like the wild-type.  Any hint of reversion would therefore be due to natural selection, and not due to some inherent tendency of a variety to turn back into its species of origin.  Reversion is therefore a poor argument for disregarding domestic organisms when studying species.

This entire conversation reminds me of work done on Atlantic salmon.  Farmed salmon are phenotypically quite distinct from wild salmon, but some of this is due to phenotypic plasticity – the farmed environment alters the appearance of salmon, without any corresponding genetic change.  When released back into the wild, farmed salmon can plastically change back to a wild-type appearance – thereby appearing to revert back to a wild state.  However, recent work has shown that this ‘reversion’ is not perfect.  Genetic changes that have occurred during farming maintain some differences between the wild and farmed varieties.  Thus domestication can create real changes to organisms that do not go away upon entering the wild.

As already stated, according to Darwin individuals within a domestic variety are more diverse than individuals within a wild variety.  Similarly, different varieties within a domestic species are more diverse than different varieties within a wild species.  Interestingly, these differences usually only occur in a single trait, the trait which the farmer finds desirable.  Therefore different breeds of cabbage differ dramatically in the leafy part, but are quite similar in their root systems; different breeds of strawberry differ dramatically in the berry part, but are quite similar in their flowers.  But the parts that are different, are different in the extreme.  For instance, from wild cabbage we have broccoli, cauliflower, Brussels sprout and kale.  If these four domestic varieties of cabbage were raised in the wild, we would likely consider them different species within the same genus, but because we know their history (we are the ones who made them), we call them the same species.  Indeed, biologists even today disagree about how many wild species are the ancestors of the domestic varieties we see.  It was only in the last few years that genetic evidence provided support for the hypothesis that all dog breeds descend from the gray wolf.  During Darwin’s time this was even more difficult (Darwin himself believed dogs to have descended from multiple species). 

And this is exactly the point: man creates domestic varieties that are so different biologists cannot tell if they descended from one species or many.  And these differences persist, being passed down from one generation to the next even when they re-enter their native environment.

Varieties of domestic pigeon, all descended from the rock pigeon (middle).  From Darwin's Variation in Animals and Plants under Domestication (1868). 

I’ve been throwing around the term variety and species as if you know what they mean.  Darwin defines them both in chapter two, but cleverly does not use his definitions for them, but rather the popular definitions.  Thus a species is a special act of creation (made by God), while a variety is a subset of a species that has been modified, such as through domestication.  Varieties therefore evolve from species, through descent with modification.  What amazes Darwin about domestic organisms is that experts cannot decide when a variety ends and a species begins!  If this is true, then perhaps species are not special creations, but rather arose through a process similar to the production of domestic varieties.  If the different breeds of dog ‘were the offspring of any single species, then such facts would have great weight in making us doubt about the immutability of the many very closely allied and natural species.’  The argument Darwin has been trying to make throughout is that variation exists in large quantities, when it arises it is passed down to the offspring intact, this variation somehow produces different varieties, and varieties somehow turn into distinct species. 

Look, for instance, at pigeons.  Darwin devotes a good deal of space discussing the amazing skeletal and behavioural differences between the different breeds of pigeon.  Each pigeon breed is quite distinct from the others, and varies not in one or two ways but in virtually every anatomic component.  ‘Altogether at least a score of pigeons might be chosen, which if shown to an ornithologist, and he were told that they were wild birds, would certainly, I think, be ranked by him as well-defined species.’  Indeed, wrote Edmund Saul Dixon, in a book on pigeon breeding, ‘My code of natural historical faith is this, that the domestic races of birds and animals are not developments, but creations’ (1851).  Darwin disagreed. Despite the startling differences between pigeon breeds, Darwin, along with many other naturalists, felt that there was ample evidence that every breed of pigeon was derived from a single species, the rock pigeon.  For instance, the pigeon breeds, despite looking enormously different, all have a similar coo, while different pigeon species have quite distinct calls, suggesting that all pigeon breeds come from one source.  Also, second generation hybrids of pigeon breeds often develop a banding pattern found in rock pigeons, a case of reversion.  Finally, all pigeon breeds can readily interbreed, whereas wild pigeon species cannot, suggesting that all pigeon breeds descend from a single pigeon species.

‘An argument, as it seems to me, of great weight, and applicable in several other cases, is, that the above-specified breeds, though agreeing generally in constitution, habits, voice, colouring, and in most parts of their structure, with the wild rock-pigeon, yet are certainly highly abnormal in other parts of their structure: we may look in vain throughout the whole great family of Columbidae for a beak like that of the English carrier, or that of the short-faced tumbler, or barb; for reversed feathers like those of the Jacobin; for a crop like that of the pouter; for tail-feathers like those of the fantail.  Hence it must be assumed not only that half-civilized man succeeded in thoroughly domesticating several species, but that he intentionally or by chance picked out extraordinarily abnormal species; and further, that these very species have since all become extinct or unknown.  Some many strange contingencies seem to me improbable in the highest degree.’

It is strange to think that a rock pigeon and a short-faced tumbler could be the same species, but our doubts vanish as we see that, by comparing breeds from different countries, we can develop a series of intermediate pigeon breeds that lead from the ancestral rock pigeon to the derived pigeon breed.

So if domestic varieties have the characteristics of distinct species, to the point that domestic varieties and wild species become nearly indistinguishable, can we really maintain the belief that species were specially created by God?  ‘When I first kept pigeons and watched the several kinds, knowing well how true they bred, I felt fully as much difficulty in believing that they could ever have descended from a common parent, as any naturalist could in coming to a similar conclusion in regard to the many species of finches, or other large groups of birds, in nature.’   Indeed, like naturalists, breeders often refused to believe that their breeds descended from a common ancestor, and they were the very ones forming these varieties!

‘I have never met a pigeon, or poultry, or duck, or rabbit fancier, who was not fully convinced that each main breed was descended from a distinct species…From long-continued study they are strongly impressed with the differences between the several races; and though they well know that each race varies slightly, for they win their prizes by selecting such slight differences, yet they ignore all general arguments, and refuse to sum up in their minds slight differences accumulated during many successive generations.  May not those naturalists who…admit that many of our domestic races have descended from the same parents – may they not learn a lesson of caution, when they deride the idea of species in a state of nature being lineal descendants of other species?’

Darwin dispels the idea that varieties are formed through crossing different species.  Crossing different species of canine would only produce a breed intermediate to the parents – it would not produce the extremes found in the bulldog or the Chihuahua.    So, if races cannot come about through mingling together different species, it must come about through the production of variation.  But variations tend to be copious, small, and in no particular direction with regard to the environment, so how are adaptations formed?  This is where Darwin’s desire to study domestic organisms becomes clear: species that are well adapted to their environment come about through natural selection, in the same way that farmers produce different breeds through artificial selection.

Domestic breeds are adapted, not for their own good, but for ‘man’s use or fancy’.  This could not be if domestic breeds were produced through changes to the external environment or through use and disuse – such things may produce different varieties, but not varieties that are so fortuitously useful to humans.  ‘We cannot suppose that all the breeds were suddenly produced as perfect and as useful as we now see them…The key is man’s power of accumulative selection: nature gives successive variations; man adds them up in certain directions useful to him.’  A farmer wants to improve on his variety of sheep, so he looks for those sheep that are the best for breeding.  His trained eye can find tiny nearly invisible differences in his sheep; but accumulating numerous insensibly tiny changes produces, through years of selective breeding, a new improved variety of sheep.

Preserved dachschund dating from early 1900s.  Notice how long its legs are - evolution of the short-legged variety  occurred insensibly such that the dachshunds of today are quite different from the dachshunds of the past.

Artificial selection by man is quite powerful, such that different domestic breeds will vary in one structure but not another.  We are interested in altering berry size, but not the leaves; the different breeds of a berry are quite distinct, while their leaves are nearly identical (not completely – correlation of growth ensures this).  Unconscious selection can also occur, which is what happened when two sheep breeders both tried to maintain the same sheep stock - insensible changes that went unnoticed over years of selective breeding resulted in the farmers having two distinct sheep breeds.  Darwin likens speciation to the formation of languages.  We start with a parental language, move through distinct dialects, and eventually have fully distinct languages, with the entire process itself being insensible.

So, Darwin says that variation is produced in large quantities (somehow, he’s not sure how); these variations are in no particular direction (rather than, for instance, increasing milk content in a cow, these variations increase milk content in some and decrease milk content in others); these variations are insensibly small, such that only a trained eye could see them; these variations are heritable, passed down to the offspring intact; and new varieties of domestic animals are formed when man selects certain varieties at the expense of others, accumulating small changes over time such that they become large changes.  That is the brunt of his argument in this chapter.

He also shows how powerful artificial selection is, by its ability to alter only the desired product.  But this leads to an interesting question: if artificial selection is so powerful, then why don’t domestic breeds turn into new species?  Although the Chihuahua could not be more different from a Saint Bernard, they are both members of the same dog species.  So what is lacking in artificial selection?  Why are two bird species nearly identical in form, and yet they cannot reproduce?

The answer, says Darwin, is that natural selection is even more powerful than artificial selection.  Whereas humans can only select organisms based on what they can see (that is, superficial external conditions), natural selection can select organisms based on slight internal differences as well, differences that could accumulate to prevent reproduction and form new species.

Indeed, Darwin hints at a concept that would not be explicated until Ernst Mayr in 1942: the importance of reproductive isolation for forming new species.  Why are there so few breeds of cat, asks Darwin?  Why do the breeds of cat tend to come from islands?  Because cats cannot be herded.  Cats roam, and breed with whomever.  Artificial selection is nearly impossible with a cat, it will breed with whomever it wants, whenever it wants.  The only way to make a new cat is to isolate it on an island, away from undesirable cats, and see what happens.  He also hints at the importance of population size.  Why are there so few breeds of donkey, but so many of horse?  Because donkeys are owned by the poor, who cannot keep enough donkeys to find those rare variants that are required for selective breeding programs.

In summary, Darwin argues that selection by man on variation produces divergent breeds adapted for man’s use.  These breeds are genealogically related, even if they appear to be distinct and even if experts often attribute them to different species.  The reader is left to wonder, if species, which are supposedly special creations by God, are so similar to domestic varieties, which are produced by the selection of man, is there really a difference in kind between a species and a domestic variety?  Or might varieties grade into species, with species being produced through natural law?