How to record EODs

Here I lay out how I record the electric organ discharges (EODs) of mormyrids and offer some advice on how you can do so yourself. This is written with mormyrids in mind, but most of this will also hold for recording EODs from gymnotiform fishes in Central & South America. What I know about EODs and recording them I've learned from Professor (now Emeritus) Carl Hopkins of Cornell University and from experience on more than one dozen field trips to Africa. I'll update this article as needed to keep it current and maximally helpful. Feel free to leave your comments or questions.

Because of their frequent species-specificity, the value of EOD waveforms in mormyrid taxonomy has long been recognized. Ideally, more researchers should be recording EODs when they collect mormyrids in the field, particularly as it is no longer so difficult to do and the size and expense of the required equipment is now rather modest. Recording EODs is especially important when species identity of the fish is in doubt and when the specimens could represent new taxa. In this article I will indicate a couple of simple ways to make high quality EOD recordings.

Mormyrids continuously produce weak electric organ discharges or EODs from an organ in their caudal peduncle. These brief (0.25-10 millisecond) electric pulses are too weak to be felt by touch (although catch a large Mormryrus and place your hand near its tail in the water and report back!), but they are easily monitored by a small amplified speaker (see article below) or recorded. A mormyid changes the interval between the pulses (called the "sequence of pulse intervals" or SPI) in stereotyped ways to communicate its intentions (aggression, submission, courtship, etc.) to neighboring fish. The waveform of an individual EOD pulse, however, remains fixed from moment to moment and its duration and shape often reflects the species and sex of the signaler (Friedman & Hopkins, 1996; Hopkins & Bass,1981). EOD waveforms are usually shown as oscilloscope tracings, or oscillograms, with time on the X-axis and voltage on the Y.

EODs are recorded by placing wire electrodes at both ends of the fish (see the earlier blog article on electrode wand construction below). The voltage between the two electrodes is amplified, recorded, and viewed. The voltage axis is arbitrary, since the distance between fish and electrodes is variable. By convention, head-positivity is always plotted upward.

Because the electric organ approximates a dipole source, positioning the electrodes anywhere near the fish will give you the same EOD waveform, although the amplitude of the EOD waveform will diminish as the fish's body axis becomes more oblique to the axis of the electrodes and will invert when the fish goes past perpendicular and switches head-tail orientation with respect to the electrodes. This is why it's very important to note the position of the fish's head with respect to the positive electrode so that head-positivity is recorded in the upward direction. (This is different in the case of gymnotiforms which have elongate electric organs, parts of which fire at slightly different times during a single EOD. The fish's angle with respect to the electrodes will affect the shape of the EOD waveform you record and in this case placing the fish in a nylon sleeve so electrodes can be fixed at the head and tail of the fish is recommended.)

One way to record EODs is to use a portable digital storage oscillscope (DSO) attached to an amplifer and electrode. However, these devices are very expensive, often bulky, designed for high-bandwidth applications and usually capture waveforms at only 8-bit resolution.The separate high quality amplifier you would need to feed signals into an oscilloscope, like the CWE BMA-200, is itself quite expensive. However, this setup would let you make DC recordings as opposed to AC-coupled recordings which filter out very low frequencies and slightly distort the waveform of very long EODs. If I were to go the oscilloscope route, I'd investigate using the USB 16-bit PicoScope 4262 ($1235) which can record 16 bit records to virtual oscilloscope software on a laptop in conjunction with an amplifier like the BMA-200.

A cheaper and more compact solution is to record EODs using consumer audio interfaces (sometimes called "sound cards") most of which now connect to a computer by USB 2.0 or 3.0. These have microphone preamps built into them which while less powerful than a BMA-200 are adequate for amplifying the EOD of a mormyrid close to the electrodes. While made to digitize signals from microphones or musical instruments, they work well for electric fish, although you're limited to AC-coupled recordings that filter out frequencies below 20 Hz and above 20 kHz, the range of human hearing. Because short EODs are just fractions of a millisecond, an audio interface with a fast digitization rate is preferable. The fastest of these devices can capture at 192 kHz (at 24 bit resolution). Not all are capable of running this fast, so be sure to get one that is. One such that is very reasonably priced ($100) is the Focusrite Scarlett Solo. I haven't tried this product, but think it's likely to work pretty well.* Its preamplifiers give you 50dB of gain. There are pricier products out there with more powerful preamplifiers, like this one. Myself, I have used an Edirol FA-66, a discontinued Firewire audio interface as well as an Echo 2 USB audio interface, also now discontinued. (The latter seems incompatible with newer laptops with USB 3.0 ports so I wouldn't recommend buying one of these.) Some new models like this one are designed to use USB 3.0 ports which supply more power than USB 2.0. 

Recording setup I used in Gabon in 2014 is shown at right. Attached to the headphone jack of the Echo2 audio interface is one of those Radio Shack battery-powered speakers, allowing me to hear the discharges of the fish. Alternatively, you could wear headphones.You will need signal capture software. I use a product called SignalScope which is only made for Mac and iOS. There are likely similar products for Windows and Android OS. I use the oscilloscope module in SignalScope. It offers all the typical controls you'd find on a real oscilloscope, including "pre-triggering" which is necessary for successfully capturing a single EOD. (If you are unfamiliar with the operation of an oscilloscope, there are a number of tutorials avialable online.) I have both the Mac OS X version and the iOS version of the software. I can also record EODs to my iPhone via SignalScope for iOS and the Echo 2 audio interface connected via a Lightning to USB camera adapter. (This also works with the Focusrite Scarlett Solo.) Yes, it's possible to record mormyrid EODs to your iPhone, although the USB audio interface will need more power than the iPhone or iPad can provide. You will need a USB Y-cable with one end connected to an external power source like a phone charger/battery pack. It's all helpfully explained here

SignalScope is pretty easy to use and saves EODs in SignalScope "projects" in which they can be looked at later and from which they can exported in text format or .MAT Matlab format. (It would be nice if future versions allowed saving notes with individual captures.)

Echo2 USB audio interface connected to iPhone 5 running SignalScope for iOSAnother option that requires less futzing in the field (but more later) is to simply record long trains of EODs with a handheld digital recorder like the Tascam DR-40. This unit accomodates XLR microphone inputs which I like for electrode cables (see article below) as well as 1/4" phono plugs. You need to record in uncompressed .wav format to get undistorted waveforms. Simply plug your electrode cable into the mic input and record the mormyrid for a few seconds at 96 kHz (its fastest setting). The preamps in this unit don't give you a lot of gain, but it's usually enough.You can open a saved .wav file in audio software like the free Audacity app. Use the magnifier tool to blow up a single pulse to see its features. While the individual waveforms will be less well resolved than individual EODs recorded at 192 kHz with an audio interface, Carl Hopkins has custom-written software that can concatenate/integrate multiple pulses yielding very nice results.

So, say you have the hardware and software sorted out and you actually have a live mormyrid, how do you actually record its EOD? 

You need yourself a basin filled with water, preferably from the capture site, large enough to accomodate the fish, obviously, and the electrode. If the electrodes are not affixed to a wand as in the model described in the earlier blog article, you want to have the + and - ends separated by a few centimeters. (No need to get fancy with the electrode: in a pinch I use 2-lead speaker wire and it works just fine.) The farther apart the electrodes are, the stronger the fish's signal will be (with diminishing returns beyond 20 cm or so). With very large fish with powerful EODs (like a big Mormyrus) it's useful to be able to have the electrodes close together to prevent overloading your recording device. In the photo above I am recording a relatively small mormyrid in a clear plastic salad container. These are great for smaller specimens since they are light weight, inexpensive and stackable. I have also used plastic troughs designed to be used planters. Larger fishes can be recorded in a cooler.

It is important to have a temperature probe inside your recording basin and that you note the temperature of each recording. Due to the physiology of their production, the duration of individual EOD pulses gets shorter as temperature increases. The change in the duration of the event for each change of 10°C is called the Q10. [A Q10 of 1 would indicate no change in rate for a change of 10°C, A Q10 of 2 would indicate a doubling of the rate. The Q10 for mormyrids have been reported to be 1.5 (Kramer & Westby, 1985) and a very similar 1.49 in Gallant (2011). Thomas et al. 1998 reported a Q10 of 1.6 for the wave-EOD-producing gymnotiform Apteronotus albifrons which has a neurogenic electric organ.] Knowing the temperature of a recording will allow one to correct for temperature differences when comparing EOD duration between individuals by means of the formula R25=RT∙Q10(25-T)/10 where RT is the time measurement at temperature T and R25 is the time measurement normalized to 25°C.

The fish need not be constrained: it can be free to swim about the recording basin. However, you need to be able to see its position with respect to your electrodes. When you make your recording you need to verify that the fish's head is towards the + electrode and the tail towards the -. Otherwise you will have recorded an EOD that is inverted relative to the standard used. (If you do so, no worries. These can be flipped in software, but you need to be aware that you've done so and note it.) ADDENDUM 25 March 2018: Often it is helpful to have a section of PVC tubing in the recording basin with the fish. A mormyrid will usually go right inside, prefrerring the darkness and cover. Often the end of its tail will be visible from one end, making clear its orientation to the electrode. Good to have PVC sections of several diameters and lengths handy. DO NOT make recordings of fish enclosed in a plastic bag placed in a basin of water. A closed plastic bag will act as capacitive filter and will distort the EOD recorded external to it.

In the field, we record the EODs of many individuals consecutively, usually fish caught the night before that have been kept in coolers or (better) separated into plastic containers or breathing bags (mormyrids kept together in coolers will fight and individuals wlll soon be missing scales from their sides).  An individual gets recorded, then placed in a euthanization basin into which an appropriate dose of MS-222 or clove oil has been added.  After its gentle death, the fish may get photographed and a DNA sample is usually taken. Then the fish specimen tagged with a unique specimen number. For this, the Hopkins Lab people (i.e. Team Mintotom) have been using an Avery/Dennison fine fabric tagging gun to affix plastic tags we print out using the printer used to print tags than go into EtOH specimen jars. With the tag affixed to the specimen (we usually insert the needle of the tagging gun into the now-dead fish's urogenital pore,  the T-bar of the tag gets locked into body cavity and minimal damage is done to the fish), the specimen is placed in a 10% buffered formalin bath to fix.

Questions? Perhaps you've tried recording EODs on different hardware/software? Share your questions or experiences in a comment using the form below and I'll add more detail where needed.


*Addendum 7 April 2018: I bought a Focusrite Scarlett Solo and have used it to record the mormyrids I keep in my home aquarium, recording EODs to SignalScope on a Mac MacBook Pro, using a microphone cable as an electrode, as described above.  It worked well, as expected. I have also successfully used it recording to an HP laptop running Windows 10 using the Sound Card Scope app. Right now, Sound Card Scope seems like the best Windows solution. I'd like to know if there us a good solution for recording to Android tablets.  ~ J

*Addendum 16 July 2019: In September 2018 we released Mormyroscope software for Windows, a customized version of Sound Card Scope by Christian Zeitnitz. See the separate article on this site.

Scratchpads developed and conceived by (alphabetical): Ed Baker, Katherine Bouton Alice Heaton Dimitris Koureas, Laurence Livermore, Dave Roberts, Simon Rycroft, Ben Scott, Vince Smith