Monthly Archives: January 2017

Recording the Voices of the Wild

Recording the Voices of the Wild – Eric Simms, D.F.C, M.A., F.Z.S., M.B.O.U.

Reprinted from British Institute of Recorded Sound Bulletin no. 2, (Autumn 1956) by permission of the British Library Sound Archive. The illustration was not included in the original article.

Yellow-billed Cuckoo (detail), via NYPL Digital Collections

Today it is generally recognized that the preservation of bird and other animal recordings in a permanent form is of importance. Not only are such recordings available to help in identification but they can be subjected to the most exact and systematic analysis. Now until comparatively recent times bird recording in Britain was carried out with the aid of disk-type recording equipment, and prior to 1939 the British Broadcasting Corporation designed and had built portable disk-recording apparatus which was entirely self-contained and powered by 12-volt batteries. These pieces of apparatus were used extensively by the B.B.C’s Mobile Recording Units, both in this country and overseas, for recording most out-door sounds, including those of birds and other animals.

However, in 1951, recording engineer Bob Wade and I began to examine the disk-type recording gear more closely in the field and to analyse the inherent disadvantages when natural history recording work was undertaken in the open air. Temperature, for example, has a direct influence on the manner in which a recording is made on a blank disk; if conditions are near freezing the emulsion that coats the disk becomes both hard and brittle and then the material, known as ‘swarf’ which is removed by the cutter as it makes the groove does not come away cleanly from the disk. This makes the walls of the grooves rather ragged and when the disk is played a certain amount of surface noise, or ‘needle-scratch’ is produced. Further, the high-pitched hiss made by contact between the playing needle and the disk is unfortunately increased by the process of taking a finished gramophone record or pressing from the original. It is, of course, necessary to make a permanent record and to ensure a long wearing quality for it; the material from which the permanent record is made must be considerably harder than the emulsion of the original direct recording.

The final records are made of a vinylite material which produces less surface noise than the normal shellac, and, after all, the permanent record is both the aim of the recording work and the determining factor as to final quality. Because of this the minimum intensity of the sound to be recorded = for example, the buzz of an insect’s wings or the call of an unborn chick in an unbroken egg-shell – must be greater than any inherent noise in the pressing. It must be remembered, too, that it is not possible to amplify a very low-level sound beyond a certain limit, since the amplifier itself contributes some noise which may be very low at the start but which increases with greater amplification. Very subdued sounds might be indistinguishable from, or unrecognized in, the general background noise coming from the apparatus itself.

For natural history recording work weight is also very important. It may be necessary to take the apparatus to comparatively inaccessible regions and to manhandle equipment across very difficult terrains. In addition to two 6-volt batteries, a motor-generator, turn-table and turn-table unit, a recording amplifier and microphones, a mixer unit to combine the outputs of several microphones, there are also the essential but nevertheless weighty blank recording disks.

Although many very good recordings of birds had been made with the disk-type recorders, the disadvantages which I have described encouraged us to examine another and newer type of recording medium which was becoming available – magnetic tape. In general, the surface-noise of a magnetic tape recording is lower than that of a gramophone record and it thus becomes possible to record and then to reproduce many sounds of extremely low intensity which are no longer masked by surface-noise . A great deal of the vocabulary of the Badger is in this category but it became possible to record all of it on tape, so obtaining low conversational calls that very few observers could ever hope to hear at a badger’s sett. The Little Ringer Plover and the Stone-Curlew were both recorded on tape and it was found that both species made calls so subdued that they could not have been audible many feet away. We have also succeeded in making recordings of the footfalls of foxes, badgers, even birds.

Early in 1951 direct comparison were made in Kensington Gardens between the two types of apparatus so far described. Simultaneous recordings were made on both disk-recorder and tape machine of the dawn and dusk bird choruses. The output of the same microphone was used so that direct comparisons could be made of the two recordings. It was found that more birds could be heard in song on the tape than on the disk, although the level of the nearest and dominant singer, a Song Thrush, was the same in both cases. The almost complete absence of noise on the former recording made it possible to reproduce more faithfully the distant songs and calls which were of lower intensity. This advantage added a natural depth and perspective to the tape recording. Of course, had these field recordings been made under ideal conditions indoors and on a high-fidelity studio disk equipment there should have been little or no apparent difference in the respective quality of the two recordings, but it is not practicable to use this type of studio apparatus in the field.

Another great advantage of tape recording is the increased duration of continuous recording that is possible with a single machine. With the single turn-table of a mobile disk-recorder it is possible to record for a period of only 4 ½ minutes before changing the disk. This would not be true of slow-speed recording at 33 1/3 revolutions per minute but this type of recording is, for various reasons, normally confined to indoor operation. There is always the chance too that the song or call might be missed during the period used for changing disks. With a tape recorder a half-hour recording can be made without changing the tape and should there be any unwanted material on the tape the operator can run this back from a second spool and by means of a third head, known as the wiping head, erase it. This allows the same tape to be used over and over again. In this way two single bird calls of some 7 seconds’ total duration were recorded over a period of 8 ½ hours; to have made certain of recording these two brief sounds it would have been necessary to cut and waste at least 125 disks. So far we have spent more than 800 hours at night making recordings of the Badger on only 25 half-hour tapes. Twelve separate calls, nine of which had not previously been described, were thus acquired but the number of disks that would have been needed for this operation would have run into many thousands.

Our present tape equipment is much less than the weight of the old disk apparatus and is far more easily moved from place to place. One type of recorded is so small that it weighs less than 20 pounds, including the spools of tape. With this machine I have been able to record massed wading birds on a remote island, the Dotterel at nearly 4,000 feet in the Grampians and the Flamingos in the middle of a lagoon in the Camargue. Tape recorders are not so susceptible to damage through vibration and their simple operation enables recording to be started at once. With the larger tape equipment it is easy to keep a continuous check on the quality of the actual recording; this is obtained by pressing a key which allows the operator to listen either to the direct sound from the microphone or to the output of the reproducing head over which the tape passes on its way to the take-up spool after recording.

In our recording work we have used the dynamic or mobbing coil microphone, and the introduction of a device known as a parabolic reflector has greatly enhanced the ability of a microphone to pick up sounds coming from the direction in which it is sighted. The moving coil microphone, without a reflector, is almost entirely omni-directional with a slightly increased pick-up of sounds coming from sources immediately in front of it. The addition of a reflector increases the effective pick-up and range for sounds coming from the exact direction in which the reflector is pointed.  Sound waves strike the surface of the reflector and are then reflected to a focal point in front. The microphone is placed at the focal point with its face directly towards the centre of the bowl-shaped reflector and its back to the source of sound. This increases both the range at which sounds may be picked up and also the amount of sound to be fed into the recorded. The reflecting surface is made of metal and the back is damped with a thick layer of sponge rubber to deaden any sounds occurring behind the reflector.

We have used two sizes of reflector for the field recording of birds and other animals; these have diameters of 18 inches and 36 inches respectively and the principles of construction and use are the same. Our increased use of these devices has opened up many new possibilities but at all times their limitations must be understood for full benefit to be derived from them. Recordings have been made at such a range that no disturbance of the birds or their areas of activity has been necessary. With the large reflector a Cuckoo in Kent with an unusual trisyllabic call was recorded at a range of 420 yards and many song-birds have been recorded at ranges of from 30 to 100 yards. Many flight recordings of birds such as Rooks, Jackdaws, and Swifts have been made at ranges of from 120 to 200 yards, and of Pink-footed Geese at a distance of nearly 1,000 yards. In addition, birds in inaccessible places such as the centres of lakes, mudflats and cliff-faces have been brought well within recording range. This extension of range has allowed us to record many birds in flight by following them with the reflector microphone; these sounds were beyond the range of the old recording apparatus.

The smaller reflector can be more easily moved that the larger and this is of great advantage when wide, uneven surfaces and cliff-faces have to be negotiated, and this portability proved invaluable when we tackled such species as the Peregrine Falcon and Dotterel in Scotland, the Flamingos in Provence and the Griffon Vulture in Spain. Snap recordings of birds can also be made without too elaborate precautions, as and when opportunity offers.

In all our work with reflectors a magnetic tape recorder has been used and this has meant the most careful co-operation between the operator of the reflector and the recording engineer. Although these new devices are improving the methods of bird-recording the original essentials of patience and determination remain. Careful field-work and study of the birds to be recorded usually means that by anticipating the actions of the birds success can often be achieved. The element of chance must always remain for it will never be possible to control the weather, the birds or the extraneous noises such as those of aircraft and trains which are the bane of the bird-recorder. Should conditions be satisfactory and the birds co-operative, the equipment is rapidly assuring the probability of success, both in obtaining a recording of the bird’s voice and in the faithful reproduction of what an observer in the field would hear.

Another piece of apparatus which has recently become available, in association with portable recording gear where the use of cables is impracticable, is a lightweight pack-set transmitter and receiver with a combined weight of only 27 pounds. This has been used with great success for the capture of certain difficult sounds, and recordings have been made of white-fronted geese feeding as much as 900 yards from the tape-recorder and of snow-buntings a mile away; it is used in conjunction with a parabolic reflector and gives the operator greater mobility, for he is no longer encumbered by the many yards of cable necessary when recording directly from a fixed microphone point.

The final recordings are, of course, obtained primarily for their broadcasting value but they are also available for detailed and systematic analysis. Today the technique of analyzing bird-songs and calls has advanced considerable. The Bell Laboratories during the war designed instruments capable by means of filters of analyzing sounds into their chief constituent frequencies; these instruments are known as sound spectrographs. Images of the sounds are reproduced on cathode ray tubes, so giving a wonderfully visual representation of the song or call. In this way simple and objective comparisons can be made between the utterances of a single individual, between those of different individuals and those of different species. Many incidental and interesting facts have emerged from such a study; a spectrogram of the song of the American Wood-Thrush showed that the performer was able to produce at the same moment of time a high-frequency trill and a low, steady tone which indicated the simultaneous action of two separate vocal mechanisms.

I have also played all the bird-recordings at very slow speed and in this way I have been able to work out the number of notes in rapid trilling deliveries during a certain unit of time. One point of interest has been the confirmation of the unreliability of the human ear in determining the number of notes given in a short period. For example, general estimates of the number of notes in the song-phrase of the cirl-bunting, which lasts for only a second to a second and a half, have been in the region of from ten to twelve. My researches reveal that eighteen to twenty three is much nearer the mark. I also discovered that the high trilling song of the grasshopper-warbler is made up of double notes of different pitch; there are approximately 1,400 of these double notes to the minute. Similarly, the river-warbler which has a song of the same type as that of the grasshopper-warbler has approximately 620 double notes to the minute. The nightjar’s song, however, is made up of single notes but of these there are something like 1,900 to the minute.

The many fields of research and experiment that lie ahead made the task of recording full of excitement, and it is a great privilege to be associated with a project of this kind.