by Tim Birkhead
There are twenty-four ornithological questions in all, but there may well have been more, for several others are scattered through the body of the Ornithology. I have arranged them into several broad categories that include identification, reproduction, migration, age and the annual replacement of feathers.
Let’s start with identification – the single most important goal of Willughby and Ray’s enterprise and still the most significant challenge of being a birdwatcher today. If you cannot identify the species you’ve seen, you cannot tell anyone about it, and if you don’t know what species you are studying, no one will take you seriously. Every birdwatcher and ornithologist will have experienced that sinking feeling when someone asks ‘I’ve seen this unusual bird; it was brown and about the size of a pigeon … can you tell me what it is?’ Such imprecision is exactly what Willughby and Ray were desperate to avoid, and why, of course, Willughby took so much effort over his plumage descriptions. If, however, someone says ‘I’ve seen this unusual bird almost as big as a cuckoo, with long wings like a martin, speckled like a woodcock … with some white feathers on each wing’ – much as Willughby’s father-in-law had done in May 1668 – we would be fairly confident about identifying it as a European nightjar.
Today we use field guides to help us with bird identification. They work because they emphasise the key distinguishing features of live birds, either standing, swimming or flying, observed through binoculars. We could say that they identify those features that identify the bird. The first truly effective guide was Roger Tory Peterson’s A Field Guide to the Birds published in 1934. Not only did Peterson excel as an artist, producing beautiful, clean, stylised images – in colour – that perfectly reduced each species to its most salient features, he also indicated by means of a short line the most important ‘field marks’ for identification.
Peterson’s ‘field marks’ are similar to Willughby’s ‘characteristic marks’ except that Peterson’s consist only of those visible on a live bird. There were no binoculars in Willughby and Ray’s day, and the telescopes that existed were so cumbersome they hardly lent themselves to observing birds.3 The only way to really see and describe a bird (or any other animal) was to examine its corpse. The aphorism, ‘what’s shot is history, what’s missed is mystery’, sums up the pre-binocular era of ornithology.4 With a specimen in the hand you had a reasonable chance of identifying it; if a bird simply flew past, you could never be sure what it was. Today’s birdwatchers, conscious of the need for conservation, are justifiably critical of that earlier approach, but they often seem to forget that it was an essential part of the ontogeny of ornithology by which the study of birds came of age.
Of course, killing a bird destroys one of its most important characteristic marks, what’s known as its ‘jizz’ – the shape, posture, mode of flight and so on. For most experienced birdwatchers today, a bird’s jizz is as important for identification as its plumage. Interestingly, Francis Willughby recognised this for as we have seen, when constructing his classification of birds, he included behaviour as one of his criteria. John Ray, however, did not.
Imagine now that I have brought you, either from a seventeenth-century Italian bird market or from my freezer at home (where I’ve accumulated – quite legally – some birds found dead on the road), three birds: a willow warbler, a chiff chaff and a wood warbler. They are all very similar in size; they are all olive green on the back and they each have a pale stripe above the eye. With no previous knowledge, how confident would you be about saying whether these represented three different species or three individuals of a single slightly variable species? This was exactly the problem facing all early ornithologists seeking to identify birds, and the only solution was to immerse yourself in the identification: to look at as many examples as possible; to measure and scrutinise their external features, noting the colour of the feathers, eyes and legs until such time that your brain shifts into categorisation mode and the birds suddenly start to form natural groupings.
I don’t think we have any idea of just how difficult this was. Perhaps only those twentieth-century birdwatchers lucky enough to explore a remote part of the world with an unknown avifauna can begin to imagine what it was like for Willughby and Ray. Even then, our modern birdwatchers would have had the advantage of being able to photograph a bird they caught to compare it with those in field guides or museum specimens, and they may also have recorded its calls and possibly taken a tiny blood sample to transcribe the bird’s molecular signature.
What would Willughby and Ray have had to go on? They may have carried with them on their travels one or two early bird books, perhaps those by Conrad Gessner and Ulisse Aldrovandi. The illustrations in these inconveniently massive tomes were poor, and they were in black and white. On the other hand, the authors’ written descriptions in both books were quite good and that must have helped. Even so, time and time again in the Ornithology, Willughby and Ray complain of the difficulty of matching up species they have examined with those described by Gessner, Aldrovandi and others whose work they quote. In addition, as we have seen, these earlier authors weren’t always especially careful and they sometimes made mistakes with both descriptions and names.
This was why Willughby and Ray decided to start afresh, seeing and describing every species for themselves, and crucially, doing so in a careful, standardised way. Once viewed from this perspective you can begin to appreciate why Willughby felt compelled to make such detailed descriptions, and in the absence of reliable colour images, to report colours in such precise ways. There were no colour standards in the 1600s (they didn’t emerge until the 1920s), so without an artist on hand to recreate the colours Willughby had no option but to compare the colours he saw on birds with those of familiar objects. Of the waxwing he says ‘The outer wing feathers are marked with spots very pleasant to behold … their appendices being red like to cinnabar or vermilion’; of the roller he says ‘the wings are a lovely blue or ultramarine colour (as the painters call it) … the head of a sordid green’; and of the dorsal surface of the song thrush, it is an ‘olive colour, from its likeness to that of unripe pickled olives, such as are brought over to us out of Spain’.5
Over the years I have worked with several bird artists who have created images for field guides. Today’s standards are high. You only have to compare the illustrations in the Collins Bird Guide (2009) with, say, the Hamlyn Guide that I used as a boy forty years earlier: the difference in the quality of both the painting and printing is remarkable. For the twentieth-century birdwatcher, those in the Hamlyn Guide are equivalent to Gessner’s and Aldrovandi’s impressionistic images of birds; those in the Collins Bird Guide are more akin to what Willughby aimed to achieve – a feather-by-feather accuracy.6 Both Willughby and the Collins’ artists, Killian Mullarney and Dan Zetterström, examined corpses – in Willughby’s case they were freshly killed, in the later artists’ case they were from museum drawers – to produce both a detailed written description and an accurate image. The main difference was that the Collins’ artists had the benefit of large museum collections that allowed them to select and illustrate a typical specimen. They also benefited from photographs of live birds to visualise the perfect alignment of plumage, which isn’t always the case on museum skins.7
One of Willughby’s questions about characteristic marks was this: ‘How many birds have an angular appendix, as it were a tooth, on each side of the upper chap of their bill, as the kestrel, the hobby, the butcher-bird, &c?’
He is referring here to a subtle feature known now as a tomial tooth, a hook on each side of the upper mandible, that occurs in two completely unrelated groups of birds: shrikes (butcherbirds) and certain birds of prey. Soon after I had written that sentence I found a dead red-backed shrike in France and was able to examine its beak and see its tomial tooth, which, I have to admit, I hadn’t previously been aware of. The fact that this structure occurs in at least two unrelated groups of birds, both of which kill vertebrate prey, suggests that it is a killing
device – an adaptation to a predatory lifestyle. Several recent studies mention the idea that raptors kill their prey with a neck bite that – facilitated by the tomial tooth – severs the spinal cord,8 but the evidence is circumstantial. In shrikes there is a small amount of experimental evidence that this is indeed the function of this subtle notch: birds whose tomial teeth had been removed made less clean cuts and took longer to kill their prey.9 Also, some other birds, such as owls and skuas that also kill vertebrate prey with a neck bite, have no tomial tooth, and a few other species also possess a tomial tooth whose function is unknown. Willughby’s question almost certainly relates to the value of this feature in systematics and probably encouraged him to position the shrikes immediately after birds of prey in the Ornithology, although this is as likely to be because of their predatory behaviour as owing to the existence of the tomial tooth. Despite these similarities, shrikes and birds of prey are not closely related, and there is no hint nor indeed any expectation that Willughby or Ray recognised what we now call convergent evolution: the occurrence of similar traits in unrelated species with similar lifestyles.
The tomial tooth is a flagship feature for the issues of classification. From Willughby’s day and before, people used anatomical features – both external and internal – as a way of classifying birds and other organisms. They had to: there was little else to go on. Willughby’s idea that you could also use behaviour or song to classify birds was not taken seriously until the early 1900s when Oskar Heinroth and Konrad Lorenz recognised and exploited its potential.10 Early naturalists assumed that God had imposed order on the natural world, and that their job was to figure out what God had in mind and reveal His arrangement. The most obvious clue was similarities in form: greenfinches and goldfinches are more similar to each other than they are to shearwaters or gulls. That much was obvious, and in theory at least it was seductively simple: one merely had to find those characteristic marks that both separated and united different species.
But it wasn’t simple at all. As we now know, natural selection has created a multitude of forms, often riding roughshod over those features that might betray the true phylogenetic affinities of a species.
In the centuries that followed Willughby and Ray’s pioneering work, ornithologists and other zoologists examined almost every conceivable external and internal feature they could think of to try to resolve the ‘order’ or classification of birds. They tried beak and bill structure, plumage pattern, eye colour, digestive tract layout (literally), the structure of the syrinx, as well as the presence or absence of rictal bristles (see below) and the occurrence of a tomial tooth, and many more features, but with only limited success. Revealing the ‘arrangement’ of birds dominated the subject of ornithology for more than 500 years and was the justification for vast Victorian museum collections.
Francis Willughby asks whether all individual birds of the same species have the same colour iris? In most cases, they do, with a small amount of variation, but in a few species the iris changes colour with age. In many gulls, for example, immature birds have a brown iris whereas in the adult it is yellow. Willughby also asks about what we now refer to as rictal bristles: ‘How many [birds] have bristles under their chin, at the corners of their mouths, or about their nostrils?’ The answer is quite a lot. They are especially prominent among nightjars, as recorded in the Ornithology, but rictal bristles aren’t a defining feature of this, or any other, group of birds. Willughby also comments on the fact that many birds have a pale rump, and wonders too whether this might be a way of categorising them. But like all those other features he and subsequent authors identified, it isn’t.
Interestingly, with all these questions, Willughby’s focus is exclusively on identification rather than function. He doesn’t ask why the shrike has a tomial tooth; why the irises of some birds – such as those of giant petrels – are pale and seem so threatening, whereas those of the giant petrel’s close relative, the albatross, are dark and seem so benevolent;11 nor why nightjars have those bristles; or why wheatears have a white rump. These are, of course, a different type of question, and one that might not have entered Willughby’s mind. On the other hand, we know that John Ray was interested in such ‘functional’ questions and spoke about them in his sermons and lectures in Cambridge, so it would be surprising if he and Willughby had not at least discussed them.12
Let us turn now to Willughby’s questions about reproduction. Two of them concern the sex ratio of birds: one is about female reproductive anatomy and one asks whether the eggs of birds ‘sometimes fall from them against their wills’.
As has been known since ancient times, pigeons invariably lay two eggs in a clutch, and Willughby asks whether it is true that these give rise to one male and one female offspring. Aristotle and the third-century ad Roman author Aelian both assumed this to be the case, and rather surprisingly, so too did the normally cautious William Harvey.13 It is an intuitively appealing idea and one that wouldn’t be difficult to promote as an example of God’s wisdom. But it isn’t true, as John Ray himself realised.
This is one of Willughby’s questions to which Ray responds. He writes that while you often do get one of each sex, sometimes the eggs produce two males or two females. The truth is that, on average, 50 per cent of the time you get one of each sex, 25 per cent of the time you get two males, and 25 per cent of the time two females. This is the overall pattern, and it is what you would find if you examined, say, 100 broods. It is a direct consequence of the two sexes being equally common at the time of fertilisation.
In birds, unlike humans and other mammals, it is the female that determines the sex of the offspring because the sex chromosomes reside in her ova (eggs). The sex chromosomes are assigned to ova at random, which is why the sex of pigeon chicks hatching from the two eggs also occurs at random. When I first read Willughby’s query about this I desperately hoped that he might have understood why Ray’s answer was correct. As we have already seen, Willughby was fascinated by games of chance, and it is chance that dictates whether you get one male and one female or two of each sex from a clutch of pigeon eggs. You can easily demonstrate this for yourself by placing an equal number of marbles (or indeed anything else), let’s say 100 each, of two different colours in a bag; remove two at a time and note down what colour they each are. Replace the marbles each time and repeat this 100 times. You will find that you have close to fifty instances of one of each colour and twenty-five cases each when the two marbles are the same colour. This is known now as the ‘binomial expansion’. However, being able to work out the sex ratio of pigeon chicks in this way, Willughby would need also to have known that in pigeons, and indeed in most sexually reproducing animals, equal numbers of males and females are produced at conception, but this wasn’t generally recognised until the process of sex determination was established in the 1920s.14
That Francis Willughby was unaware of equal sex ratios among birds is clear from his next question. ‘In what kinds of birds are there more cocks usually bred, as in ruffs, in what more hens, as in poultry?’ The ruff is a wading bird that was once abundant in the marshy regions of Lincolnshire and may well have been among the species Willughby saw there in the summer of 1662. Ruffs were excellent eating and local Lincolnshire fowlers ran a cottage industry capturing and selling them. The birds were caught in clap nets, kept in the dark, and fed on a diet of bread and milk until they were fat enough for the table. Keeping them in the dark prevented them from seeing, fighting and killing each other. Willughby’s Latin name for the ruff, Avis pugnax, says it all: the pugnacious bird. At least, the males are aggressive. The ruff has an unusual breeding system in which males congregate at a communal display arena (referred to as a lek) where they compete aggressively for a tiny patch of ground that provides them with the opportunity to copulate with females on their visits to the lek. After copulating, the female – known as the reeve – goes off to incubate her eggs and rear her chicks alone. Most of the time, therefore, more males than females are pr
esent on the lek, and hence when the Lincolnshire trappers caught ruffs at their leks they generally secured more males than females, giving the impression that the sex ratio was strongly biased towards males. The intense competition between males for matings on the lek has resulted in their evolving both aggressive behaviours and also elaborate head and neck plumes – the ruff. Because it was once believed that wild populations of ruffs contain more males than females, there have been several attempts to assess the ‘real’ sex ratio of ruffs. Contrary to expectation, these all indicate a slight preponderance of females in the adult population.15
Willughby’s idea that in poultry – by which we can presume he means domestic fowl – females outnumber males is also a consequence of their mating system. Typically in domestic fowl, and in their wild ancestor the red jungle fowl, a single dominant cockerel assumes control over a harem of several females, driving away the younger, more subordinate males. But the sex ratio among adult fowl is close to even. Indeed, if the modern poultry industry had found a way to break the biological fifty-fifty stranglehold of this aspect of avian reproduction, they surely would have done so because the profit is largely in hens rather than cockerels.
This fascination for sex ratios may have had its origin in Venice when Willughby and his colleagues were there, discussing the slight excess of human males revealed by the 1581 census. Ray comments that this is similar to what has been found in London: ‘And I doubt not but if exact observations were made in other places, there would be found the like [same] proportion between the numbers of males and females born in the world in hot countries, as in cold, so that from this topic, the Asiaticks [sic] have no greater plea for multiplicity of wives, than the Europeans.’16