Part 5: Hamish taught me all he knows about bird sounds

The egg

I am still amazed that ducklings in the egg have been shown to be able to hear their mother for a week before they hatch. Once the eggs are pipped and air can get in, Goethe (1937, 1955) discovered that Herring Gulls will call from the egg when the colony is disturbed or when the egg is shaken. From studies of Mallards Anas platyrhynchos, Eurasian Skylarks and other species, it has been shown that many young birds call to each other in the egg and, where it is to their advantage, birds even synchronise hatching by calling to each other (Vince 1969). 

Call development

Chicks stimulate their parents to feed them with calls, and in many species young can distinguish their parents’ sounds from those of other adults. Sandwich Tern chicks for example start calling when they hear their parents calling above the colony. Rosemary Jellis described work by Tschanz (1968) who showed that young Common Murre Uria aalge recognise their parents individually by call. If their own parent doesn’t return from the sea after a day or two they seek warmth and food from another adult, stimulating adoption by learning and responding to the new adult’s calls.

Using sonagrams, Thielcke (1976) illustrated beautifully the way that begging calls of young Short-toed Treecreeper Certhia brachydactyla slowly change in structure over the course of 25 days, until they become adult social calls. He also showed how the adult song consisted of calls arranged in a learnt set sequence and that each sequence was characteristic of a Short-toed population.

In the late eighties I started to concentrate my birding on Poole Harbour, an area of nearly 100 square miles of beautiful scenery and great birds. Around the same time, researcher Robin Ward found a Wilson’s Phalarope P tricolor in Holes Bay, just outside where I was working. Robin had come from Wales and was surveying for an environmental impact study. He suggested that we started a bird pub to encourage Poole birders to come together, discuss what they had seen and learn more about birds. Over the next 15 years the Poole pub idea flourished and I made some great friends there. In the nineties one of the various puzzles we all discussed was that of autumn Common Chiffchaff call development. Around this time we were all getting very interested in the idea that the chiffchaff could be split into several species, and were listening more carefully than we had before. For two or three autumns, many of us noticed that instead of the normal huit we were hearing wheeoo calls (CD1-86). The suggestion was that chiffchaffs had changed their common call. Was it a different population that was now passing through, and might they even represent a different taxon that had previously been overlooked?

I had read in BWP that young Common Chiffchaffs (of the nominate subspecies P c collybita) had different calls. Then one autumn day, while birding with Paul Holt on Ballard Down overlooking Poole Harbour, the conversation turned to juvenile calls. We heard a chiffchaff giving the anomalous wheeoo call and Paul commented that he also thought that this was the characteristic call of a young bird. After this autumn, however, birds calling this way seemed to stop showing up: there were very few for several years. What had happened? Surely, if they were first-year birds we would hear lots of them every autumn?By this time the subject had started to intrigue quite a few birders, and Bill Oddie (1997) even suggested in his ‘Gripping yarns’ column in Birdwatch, that we might be witnessing the birth of a new species. I discussed the likelihood of them being something exotic with Martin Cade, the warden of Portland bird observatory, Dorset, who was catching 100s of chiffchaffs. He told me that several of the birds which had called wheeoo upon release had shown all the in-hand characters of the nominate. Killian then clinched it for me when he took a dictaphone into his garden in Wexford, and recorded an adult feeding young giving wheeoo calls.

The debate was still not over, however, and whenever I tried to argue the case for the sound being that of a young bird I could never explain why the calls were now heard far less commonly than in some previous autumns. Then it occurred to me that perhaps the wave of chiffchaffs we encounter at coastal migration hotspots in Britain might differ in their stage of vocal development from year to year. The autumns when we are flooded with chiffchaffs using wheeoo calls could be those where there are large numbers of late broods, either caused by poor springs on the breeding grounds or, alternatively, by a favourable breeding season resulting in large numbers of second broods. Other years, chiffchaffs might be a little older by the time they pass through, with most of them already giving adult-like hueet calls. When we age birds we do so mainly by plumage. Usually, this relies on a very straightforward set of criteria, and when a songbird is moulting it typically does so either before migrating or afterwards, when it has arrived at its wintering grounds. Unlike moult, call development has no bearing on the ability of a bird to migrate. Until this point I had failed to realise the obvious, that these two kinds of development could proceed independently, each at their own pace.

Listen to a recording of a Common Chiffchaff at a very early stage of development (CD1-87), a juvenile on the 3rd of August. At this stage, calls of the nominate can sound remarkably similar to those of adult Siberian Chiffchaff (tristis), as you can see in the sonagram (compare with the Siberian Chiffchaff of CD1-15). These calls are typical of the first couple of weeks after fledging, so they shouldn’t be a problem in places where tristis turns up in October. Returning to CD1-86, which was recorded during autumn migration on the 9th of September, most of the calls in it could be described as wheeoo, but a few are less inflected (eg, 0:16-0:18), and reminiscent of the early stage you can hear in CD1-87. A call of an adult Common Chiffchaff from CD1-13 is shown for comparison.

As you now hear, sound development is not straightforward, and can appear to move forward and back. A practical aspect to this is that in spring, both adult males and females will sometimes use immature calls. In chiffchaffs, for example, immature wheeoo calls can sometimes be heard in the spring, quite often interspersed with normal huit calls. Adults use immature calls in courtship in the same way we may use baby talk when trying to be endearing. We don’t have examples of Common Chiffchaff doing this but we do have these calls given by an adult female Mealy Redpoll A flammea in north-eastern Siberia begging for food in the company of two adult males (CD1-88a). You can compare them with calls of a begging juvenile in Finnish Lapland (CD1-88b).

Learning to sing

During another Poole bird pub session in August 1995, I was asked by ringer Dave Dicker why, that day when he had ringed a Cetti’s Warbler he knew to be a fresh juvenile, it had flown to a song post and sung. I didn’t know the answer, but I looked it up in BWP when I got home and was shocked to read of an endless array of passerines that were known to sing at as little as six weeks old (Goldcrest and Firecrest even start at four weeks). Here is the song of a young Cetti’s (CD1-89), and an adult to compare it with (CD1-90).

That winter, while listening to Songs of the warblers (Borror & Gunn 1985) as I attempted to learn American warbler songs for the World Series of Birding race in New Jersey in May, I read what has to be the most comprehensive set of sleeve notes ever. In a few paragraphs it was explained why the sounds I could hear in New Jersey’s Cape May woods were not always those on the tapes. “By the time they reach Canada’s southern boundaries in spring, most warblers that nest in the north are singing their primary songs. This is not the case, however, in much of the United States, where many warbler species are seen only on migration, and may not have perfected their primary songs.” 

These things had always interested me, and over the subsequent years I learnt that the stages of song development I had read about in ‘American warblers’ also apply for many European bird families. I also discovered that in the earlier stages, songs could be produced by both sexes, although after a few months, at least in many species, the females stop singing. 

I suspect that there is a great deal more for scientists to learn about the strategies behind song development. The learning process leading to adult song is highly flexible, and has several identifiable stages. At present it is thought that the first of these is subsong, where there are hardly any recognisable features of adult song. Gradually, rudiments of adult song start to fall into place, but in sequences that are far longer and more rambling than in adults. At length, often as late as early spring, a songbird progresses to a stage called plastic song, in which songs gradually shorten and phrases begin to become more stereotyped. Songs finally crystallise when the breeding season gets going. But even after this, some species continue to learn, in some cases adding new material to their songs for many years.

Subsong

Nicholson (Nicholson & Koch 1936) was the first to use the term subsong to describe song of low volume that doesn’t seem to have any relevance in territorial arguments, or invoke any hostility in other males. Subsong has great potential to confuse the birder, because it is usually given from dense cover, is often full of mimicry, and may bear little resemblance to familiar adult songs we usually rely on in species identification. It is most common in first-year birds. Listen for a continuous stream of twittering, stammering and babbling, littered with call notes, typically on August and September mornings.

Nice (1943) studied Song Sparrows Melospiza melodia as they progressed through various stages of vocalisations. Her descriptions were then defined by Thorpe (1961). The details vary between species but on the whole his description has not needed much improvement since then. Thorpe pointed out that subsong is quieter, with an entirely different pattern of notes and it includes more call notes and mimicry than ‘normal’ song does. Song bursts tend to be longer, and the frequency range varies more than in the full song. Subsongs are typical of birds with a low sexual motivation, for example adults and first-year birds before the breeding season really gets started, or juveniles after it has finished. Listen to autumn subsong of a Common Blackbird hidden deep in a berry-laden Elder bush (CD1-91).

Most of us do not pay much attention to these songs, not really understanding what they are, and perhaps preferring to ignore them, like ducks in difficult eclipse plumages. Sometimes, however, like when thrushes tune up in this way in early spring, they can create some confusion as these quiet, warbler-like songs from deep inside a bush are mistaken for, eg, Marsh Warblers. 

Plastic song

Plastic songs are clearly recognisable and can be identified to species as long as these have identifiable song phrases. However, the timing will often seem wrong and a bird will run phrases together and then sing normally before dropping back into subsong. Plastic song is louder than subsong and the individual phrases are shorter. Fragments of fully formed adult song are typically intermingled with gentle immature babbling. The timing during the bird’s calendar, its age, is important and birds often start to crystallise their songs after a winter break. Then, a young bird may be spending a lot of its time singing. The term plastic song is derived from the pliable or ‘plastic’ nature of the notes and motifs in it. 

Common Chaffinch has been the subject of much of the research on vocal development, including plastic song. This species begins to resolve the finer details and add a terminal flourish to its song around the time it is establishing itself in its own territory. A Common Chaffinch song without a clearly defined terminal flourish will most likely be a first-year bird singing plastic song. The process of crystallisation gradually takes place under the influence of the songs heard from its new neighbours. Listen to the recording illustrating Common Chaffinch plastic song (CD1-92) made by Dick on a sunny day near the migration watchpoint of Hanko. In most of Europe we take Common Chaffinches for granted but in Finland they are spring migrants arriving in places like Hanko in large numbers. Although it was sunny, the temperature was -10ºC.

Common Nightingale illustrates the three stages well. Its subsong (CD1-93) is quieter and has an entirely different pattern of notes compared to adult song; it also includes more call notes than a normal song does, including lots of chattering ones, as well as a short whistle near the start, and rattles towards the end. In the case of nightingale, subsong doesn’t include more mimicry because adult nightingales already include so much in their songs. Neither are the song bursts longer, but when he described subsong, Thorpe was writing about Common Chaffinch in which the normal song bursts are shorter. The frequency range of the nightingale’s subsong is greater than usual in normal song. 

Now progress to plastic song (CD1-94) which unlike Common Nightingale’s subsong does fit the textbook description.  You can hear the song starting to crystallise as it progresses from subsong. The timing isn’t right and the phrases run together, then it does sing normally before dropping back into subsong. This plastic song is louder than the earlier example of subsong and the individual phrases have become shorter, and fragments of fully formed adult song are intermingled with gentle immature babbling. This could be mistaken for Thrush Nightingale. 

Finally listen to a beautiful, fully crystallised song of Common Nightingale, recorded by Arnoud near his home in the Netherlands. It includes long gaps between phrases, allowing the singer to listen to responses from its neighbours (CD1-95).

Jackie and Nick Hull are two regulars at Poole’s bird pub. Early one spring, I told them where I had heard a Garden Warbler S borin singing deep in some undergrowth. Seeing myself as a bit of a whiz on bird sounds, it was pretty embarrassing when at the next pub evening Nick told everyone that it had turned out to be a Blackcap S atricapilla. In England this mistake is frequently made, but why? And why, in my case, could I consistently identify Garden Warbler song and once even managed it from the open window of a speeding car, and yet made this kind of mistake with Blackcap. Curious, I started to ask every birder I met how they separate the songs of Garden Warbler and Blackcap. Some answers were best summed up by Percy Edwards’ description of the Garden Warbler taking singing lessons from a brook passing under a small bridge, and his comparison of Blackcap to Pan running his bearded lips over his seven pipes before bounding through the woods. Svensson, ever the poet in the Collins bird guide, had described Blackcap “turning into clear melancholy flute notes at the end”.

A few of my experts joined Percy and Lars and identified the Blackcap songs by timbre. “Garden Warbler has a mellower, blackbird-like timbre” said Magnus for example. Others used pitch: “Garden Warbler constantly drops into the lower notes while Blackcap reaches for the higher”, but most used the length of the song “Garden Warbler sings appreciably longer songs than Blackcap”. One thing I noticed in probing was that Scandinavian birders did not see the Blackcap/Garden Warbler conundrum as much of a challenge at all. 

This gave me a clue and I started to realise that, as with migrant warblers in North America, many of the Blackcaps we hear in Britain and southern and western Europe in the early spring are wintering birds or are still en route, so when we hear these migrants they are singing subsong or plastic song or a mixture of both. 

In Poole Harbour where the mistake was made, Blackcaps winter in gardens in reasonable numbers, and ringing recoveries have shown that in late March and early April they leave to breed in Belgium and Germany. Meanwhile other migrant Blackcaps pass through the area from Africa, singing as they go. Simultaneously our local breeding birds arrive and set up territory. Garden Warbler, on the other hand, winters in Africa and is a less common spring migrant and much rarer breeder.

As a consequence, the picture for Blackcaps as they move through parts of Europe is complex, for as they travel they also develop through the full range of songs, crystallising their songs when reaching the breeding grounds. Meanwhile, the Garden Warblers singing the songs I identified from speeding cars are African migrants that have arrived directly at their breeding grounds, singing nearly crystallised songs. 

As described by Thorpe, subsong and plastic songs, timbre, prevalent pitch and length of the song phrases are wonderfully varied, making it impossible to use any of the normal identification criteria. Here as a guide are three different stages in the development of Blackcap song. Then these are compared with crystallised Garden Warbler songs.

One of the problems of recording subsongs is seeing the bird, as subsongs are normally performed by birds hidden deep in a bush. Magnus didn’t see the bird he was recording (CD1-96) and it’s one of very few tracks in this guide where the bird hasn’t been visually identified. Still, he did see a Blackcap subsequently 

and I can vouch for it sounding just like the 10 or so Blackcaps that winter in my garden and entertain me by singing subsongs out of sight all through the early spring. Again this subsong is quieter than the ‘normal’ song, with a different pattern of notes, and the bird stops to call for a period in the middle. It is difficult to catch any mimicry but that doesn’t mean it’s not there. The song bursts are longer, and the frequency range varies more than in the full song. Although it is difficult to prove, this subsong has characteristics that suggest that it was sung by a female, for example the ‘weeping’ call (cf Bergmann & Helb 1982).

As Blackcap songs crystallise they go through several stages, and this next recording is an example of plastic song or possibly even what Marler & Slabbekoorn (2004) call early plastic song (CD1-97). You can hear the crystallisation process in action as a loud whistled adult-like phrase emerges from the chatter of the subsong after 10 seconds. There are hardly any gaps in the song, which lacks clear phrasing, and much of the time it strongly resembles subsong. 

In studies of song development the full crystallisation of song seems to happen suddenly, and this final Blackcap example (CD1-98) is a full song recorded in an idyllic thatched cottage garden in the heart of Dorset. We have also included the song of an adult Garden Warbler for you to compare these Blackcap songs with (CD1-99).

Another Sylvia warbler that sings as it migrates is Lesser Whitethroat and in Dorset, towards the end of April, I pointed out a singing but hidden one to my wife, Mo. On returning home, I wanted to show her the ease of using The Sound Approach database. I looked up and played an example of a Lesser Whitethroat song from our collection (CD2-01). When she didn’t recognise the bird in the recording I then found another recording made in late April of an early plastic song, the intermediate stage between subsong and plastic song (CD2-02), which she immediately recognised as the song she had heard that morning. 

I had mentioned in my original paper (Constantine 1994) that while you are attempting to learn calls and songs, so too are some of the birds you are listening to. What I hadn’t mentioned is that in order to compare, you should always rely on referenced recordings where the date and location are given. This increases the chances of you comparing birds of the same age and stage in song development. Papers on bird identification have been doing this with photographs for years, but it also has to become general practice in discussions about sounds. 

An awareness of plastic song is important when identifying spring rarities, many of which are first-year birds. In species where first-year males normally return to where they hatched, songs of vagrants are likely to be immature; a vagrant is by definition outside of its normal breeding range, so it won’t have the benefit of feedback from other males to help crystallise its song. On the 23rd of May 1994, Alan McCall found a warbler in the Kergold plantation in Shetland and he thought that it could be a Marsh Warbler. The next day another birder thought it was a ‘Hippolais’ warbler, then the concensus later swung back to it being a Marsh Warbler. On the 26th of May, Bill Jackson recorded it. It was singing a plastic song or as Bill put it “quiet much slower melodic and fluty”. It was eventually identified as a Blyth’s Reed Warbler. He asked me at the time to analyse the songs, which I did without reference material. Now thanks to The Sound Approach we can compare the two species at this stage in their development. Plastic song of spring migrant Blyth’s Reed (CD2-03) can be much faster and more varied than fully mature song heard on a breeding territory. As we have learnt this could lead to misidentification as Marsh Warbler, whose plastic song is faster and more complex at this stage in its development (CD2-04). 

Not everything young sounds so different. Fan-tailed Warblers are capable of producing a very adult sounding song at a very early age. Every summer in the south-western Netherlands, from early July onwards, Zitting Cisticolas Cisticola juncidis turn up and start to sing as they set up new territories (CD2-05). These are thought to be juveniles dispersing from the nearest regular breeding grounds in western France (van den Berg & Bosman 2001). Given the very short time they have before claiming their own territories, it is not surprising that the song of these birds is slightly higher pitched and sharper. This difference is also noticeable when the birds first sing in the spring. 

I have to admit that once Shaun Robson, another Poole birder, had got the idea of song plasticity he did start to find birds singing these songs. I was a little gripped when he heard what must have been a juvenile Common Grasshopper Warbler because I had always wondered whether a juvenile Common Grasshopper or Savi’s Warbler could be mistaken for each other. I was thrilled when we recorded this juvenile Savi’s (CD2-06) and this juvenile Common Grasshopper (CD2-07) singing in mid-August. Common Grasshoppers also produce a ‘wobbly’ kind of plastic song (CD2-08) and it can sound like a different Locustella species; maybe a little like Lanceolated Warbler. 

Go to the right habitat on a calm morning in late summer, especially when migrants are passing through, and you will have a good chance of hearing subsongs. They are very pertinent in the identification of Phylloscopus warblers. By sound, an increasing number of Siberian Phylloscopus warblers are being found wintering in western Europe, although you could doubt your ears if you had listened to commercial recordings of adults on the breeding grounds. For many of the rarer ‘phylloscs’, breeding as far away as Siberia, these are the only kinds of songs we are likely to hear from them. Listen to this Yellow-browed Warbler subsong recorded in late September (CD2-09). The date alone would suggest that this is what you would hear, and only the first hints of adult song are starting to emerge. As another example listen to a recording of plastic song of a Pallas’s Leaf Warbler recorded in early April near Rotterdam, the Netherlands (CD2-10). Again the date and place would lead one to expect plastic song and it sounds as if the bird is ready to crystallise its song in competition with other males. However, a vagrant like this may never have the crystallising opportunity and could stay at this stage throughout the spring, or develop some kind of mixed song. 

Common Chiffchaffs are usually the only Phylloscopus warblers to be found in winter in Europe, but every year a few Siberian Chiffchaffs are discovered among them. Given the difficulties of separating Siberians from other chiffchaffs on plumage, it is useful to be able to recognise not only their calls (CD1-15) but also the kinds of songs they may give from autumn through to spring. Listen to a spring migrant Magnus recorded in Almaty, Kazakhstan, on the 16th of May. He captured this bird in the final stages of crystallisation and in the first recording it is singing plastic song, usefully littered with call notes (CD2-11). In the second cut you can hear how its song is structured. It now sounds much as a Siberian Chiffchaff will sing upon reaching its breeding grounds further north, still frozen as this bird sings in May (CD2-12).

Unlike many songbirds, which mimic other birds more in their subsongs than in their adult songs, Marsh Warbler does this less in its subsong (CD2-13). Luckily, even at this very early stage the difference in tempo between this song and that of European Reed Warblers (CD2-40) is already apparent, Marsh having the faster and more varied song. The European Reed’s subsong in CD2-14 is much more like the adults’ crystallised song. 

Sometimes it can be hard to separate the songs of Common Blackbird and Mistle Thrush T viscivorus. At the beginning of the season, when their songs first start to crystallise, blackbirds sing simpler songs than later. In areas where they are partial migrants, some stay to winter in towns and are in position to breed early in the season. In the towns, it is warmer, there is plenty of food, and population densities can be very high. As the weeks pass and competition increases, the songs of blackbirds breeding at higher densities become more complex and easier to separate from Mistle. Those blackbirds that flew south for the winter return to breed when their urban relatives’ breeding season is well underway. Sometimes these birds choose traditional rural sites where they breed at much lower densities. There is not as much competitive pressure to enrich their repertoires as there is in town, and consequently the songs of these forest blackbirds remain simpler throughout the breeding season. The situation is similar for Mistle Thrush: listen to one which I recorded from the back of a frozen Finnish chalet where it had only just arrived. The majority of Mistle here were still in flocks foraging along the frozen tracks, and this song had to be edited to remove the sounds of Dick singing as he ran out of the sauna and into the snowy air (CD2-15). Now compare it with one at an earlier date but a later stage in a much longer breeding season in Scotland where Magnus was thinking more of warm porridge than saunas (CD2-16).