Not very familiar with sound production in insects, but the fundamental sound frequency and harmonics should give you cues on the physical structure producing the sound. The harmonics should be related with the resonance frequencies of the biological structure involved in the sound production.

There are definitely clues present in spectral components of signals you recorded, but for unambiguous confirmation, anatomical analysis will be necessary. For example the frequency can tell you whether a frequency multiplier (such as the scraper-file combination) is involved, since insect muscles cannot contract faster than 1 KHz. This paper gives an overview of some of anatomical basics of sound production in insects:

Michelsen A. & Nocke H. (1974). Biophysical aspects of sound communication in insects. Adv Insect Physiol 10: 247–296.

Is it possible to see a representative spectrum and/or spectrogram of what you have?

Hi Jernej, nice to have your answer here: you are in the same institution of my friend (and teacher) Paolo Fontana.

Here attached is a screenshot of a song.

My field is vibrational communication, so I don't have much experience in interpreting grasshopper sounds, but I think this non-resonant spectral structure could be produced by striking. Since your animals don't have a specialized stridulatory structure, try looking for a more general motor pattern, such as the wings hitting the abdomen or each other. It might help to acquire a high-speed camera for recording the animals while singing, then look for a moving body part. Once you identify the structure that moves, you can do a detailed anatomic analysis. Maybe the exoskeleton is thickened at the place of the strike, or something similar. But I'm already guessing.

I think that Jernej's camera suggestion is the best solution. Even a $350 Fuji camera equipped with a zoom lens and a simple macro lens clipped to the front can take video at 320 frames (or more, depending on the model) per second. The resolution is not great, but it will definitely be enough to allow you to resolve anything oscillating at less than the Nyquist frequency -- in this example, 160 Hz, or one-half the "sampling" (= frame) rate. We have done this recently with singing green lacewings, with good success.

I also agree with Jernej that the sonogram looks like percussion of some sort.

Thank you all for your suggestions. I have video recordings, taken with a digital photocamera, to check. I have two hypotheses: one is that this grasshopper use some body part against the substrate, the other is the involvement of tympana (something like cicada tymbals). I look at the video recordings and let you know if something get more clear.

Let me know! -- I'm very interested. Sound production via tymbals/tympana has not been reported before in grasshoppers, so that hypothesis might not be worth pursuing too far. But who knows? Insects always surprise us!

I was very surprissed about the alar-notal method sound production. The song is produced by rubbing together the ventral edge of the metanotum and the basalar sclerite. You should read this interesting paper: Sound production mechanism in pamphagid grasshoppers (Orthoptera). Journal of Zoology 275 (2008) 1–8.

Sincerely. Pablo Barranco.

Sound production mechanism in Pamphagidae grasshoppers (Orthoptera). Journal of Zoology 275 (2008) 1–8.

Thanks Pablo!

The paper (see attachment) you mentioned is very interesting, I will check in my dry specimens of Netrosoma if there is similar structure for sound production.

I will let you know, thanks again for calling my attention to this publication.

Ciao

Filippo

I agree with Charles Henry about anatomical validation. In our lab we have modified a GoPro action camera (which records at 240 frames per second and fantastic resolution) with macro lenses and have been getting incredible "high speed" video of echolocating bats for a fraction of typical high speed camera costs. Contact me directly if you'd like more information on this setup--all the components are available commercially from various companies.

I've had some success isolating sonic musculature and swim bladder output in fish, is that kind what you're thinking? based on the timestamps I'm guessing you already published, I'll have to work backward on this Q&A...

I don't think it's straightforward to infer the nature of the physical noise-making structure from the characteristics of the sonogram/spectrogram.  At least, that seems to be the case in insects.

However, I would be very interested to hear from anyone who has additional insights about this question!

Best wishes, Charlie

Tell you what - could you send me reference file? If it's classic impulse impulse response behavior we'll see it right away.

You won't be able to determine anything absolutely without observational (physiological data). One crazy possibility if the mechanism is similar but no file, would be a slick-stick mechanism as shown by Sheila Patek in arthropods and us in catfish (Mohajer et al) where instead of individual collisions the wings would rub, stop and produce sound when they exceed static friction. I have no idea if this is a real possibility or not.

I completely agree with Michael, let me revise my pitch - I expect if you have a priori understanding of sound generating mechanisms AND some audio files, we can infer the sources from the audio files alone. Worth trying anyway.

 in some special cases waveforms and spectrograms can help in understanding the mechanisms behind the sound production. In some cases of ants and coleoptera we clearly see that sounds are made of series of clean discrete pulses in sequences clearly generated by a movement of a plectrum against a well structured pars stridens, also showing if the sound is produced in alternated movements or not (e.g. by looking at the phase of pulses). If multiple parts work together or if resonant structures are involved, then the image produced by spectrograms can be much less clear. There are other cases where spectrograms show features useful to understand the sonic mechanisms, but no general rules can be drawn. Often it is also important to set the correct spectrogram and visualization parameters to see what we need.