The cactus deermouse sings

Cactus deermouse

Awhile back, I wrote a blog post about singing mice.  I’ve been wanting to learn more about this topic every since, but was stymied by the lack of mice in my yard.  This winter my veggie garden was overwhelmed by rodents that seemed determined to make sure I never got a taste of fresh broccoli.  I tried repelling the hungry monsters with odiferous substances, and then tried fencing them out, but apparently they really, really like fresh broccoli, lettuce, and peas.  I finally set up live traps, and ended up relocating 7 packrats, 6 round-tailed ground squirrels, and a couple of mice.  I caught two more mice, but put them in cages to see what they had to say.  Yes, I do understand that most people might think it odd that I was trapping mice outside to bring inside.  But what some might call odd, I call science.  One mouse was a harvest mouse (Reithrodontomys, probably megalotis), and the other was a cactus deermouse (Peromyscus eremicus).  Both were females, and the cactus deermouse looked quite pregnant.

Fortuitously, just about the time I captured the mice, a new app became available for Android phones for recording using certain ultrasonic microphones.  The app developer, Bill Kraus, asked if I would beta test, which I was happy to do.  I hooked up my Dodotronics Ultramic to my Samsung S3 (and later my LG G4), and set up to record.  The great thing about the app is it shows a live spectrogram, so you can see what is being recorded.  It also has a feature for stimulus triggering, so it starts recording when it “hears” sounds of user-set frequency and loudness.  To learn more about recording ultrasounds with your phone, see here.

I started with recording the harvest mouse.  She basically had nothing to say, and the only vocalization she made during several nights of recording was an alarm call when I accidentally startled her.  Other studies have also found that harvest mice don’t vocalize much, except for the Mexican harvest mouse (R. mexicanus), which is found primarily in wet montane habitats from Tamaulipas to Ecuador.   Interestingly enough, the champions of the singing mice world, literally known as singing mice (Scotinomys spp.), are also found in montane environments in Central America.  Is there something about montane tropical forests that promotes singing in mice, or have mice just not been studied enough to determine if singing is really limited to these habitats?

I then moved the microphone to the cage with the cactus deermouse.  Female rodents are known to kill their young if they get too stressed, so I was very careful not to disturb her – I didn’t even change the litter in her cage.  I gave her fresh food and water every day, but otherwise let her hide in the nest that she built.  I was pleased to see that she actually vocalized quite a bit!  Her calls were pretty simple, a string of 3-7 syllables, all in the ultrasonic range of 20-32 kHz.  She seldom came out of her nest until after I went to bed, so I didn’t get to see what she was doing while she was making these sounds.  The recordings below were both lowered in pitch and slowed down, so you can hear the sounds.

Spectrogram of calls of a female cactus deermouse. Frequency scale on right goes from 0 to 100 kHz.
Spectrogram of calls of a female cactus deermouse. Frequency scale on right goes from 0 to 100 kHz.

One evening, while I was reviewing the recordings on my phone, I noticed some new sounds on the spectrogram – much higher pitched than what I had recorded previously.  Just then, I heard a noise from the cage, and looked over to see 3 pairs of mouse ears pointed in my direction.  The babies were out!  There were 4 total, and they were about half the size of the mother.  The babies were not as shy as their mother, and started appearing every night around 9 pm.  In addition to eating, they also played a lot, chasing and wrestling, and climbing all over the sticks I had placed in the cage.

Poor photo of one of the baby cactus mice
One of the baby cactus deermice.

The recorder picked up a lot of their vocalizations also.  These were higher pitched and covered a large frequency range, from 30-75 kHz, with most of their calls above 50 kHz.  Their calls were also very fast, with each syllable just a few hundredths of a second long.  They were also much more variable in form – although most of the calls descended, some ascended, and some did both.

Spectrogram of the captive juvenile cactus deermice. They are about 3 weeks old. Scale on right goes from 0 to 100 kHz.
Spectrogram of the captive juvenile cactus deermice. They are about 3 weeks old. Scale on right goes from 0 to 100 kHz.

With all 4 juveniles in the same cage, it was difficult to tell who was vocalizing, or in what circumstance.  I achieved my primary goal, which was to determine if cactus deermice vocalize.  They obviously do, and their vocalizations appear to have some of the structure of songs.  I also established that juveniles vocalize differently than adult females do.   Not wanting to stunt the development of the babies, and not having a better research setup to answer the myriad follow-up questions raised by my observations, when the babies were about 4 weeks old, I released the entire family back into the desert.

A few other members of the cactus deermice genus (Peromyscus) have been documented singing.  Both male and female California deermice (P. californicus) and brush deermice (P. boylii) sing during the mating season.  It’s probable that most species of the genus (and there a lots) do some form of singing.  Many of these species are sympatric (live in the same area), so vocalizations likely differ among species, and may even be used in species recognition.  In addition to being ultrasonic, thus inaudible to human ears, the vocalizations are not very loud.  Ultrasonic vocalizations don’t travel very far, and are easily deflected by vegetation.  So studying mouse vocalizations in the wild is a difficult task, but as equipment gets more refined, I think we have a lot to learn about communication in the rodent world.


Sales, G. and D. Pye. 1974.  Ultrasonic communication by animals.  Chapman and Hall, London.

Campbell, P., B. Tasch, J.L. Pino, O.L. Crino, M. Phillips, S.M. Phelps. 2010.  Geographic variation in the songs of neotropical singing mice: testing the relative importance of drift and local adaptation.  Evolution 64:1-18.

Kalcounis-Ruepell, M.C., J.D. Metheny, M.J. Vonhof. 2006. Production of ultrasonic vocalizations by Peromyscus mice in the wild.  Frontiers in Zoology 3:3.

Miller, J.R. and M.D. Engstrom. 2010.  Stereotypic vocalizations in harvest mice (Reithrodontomys): Harmonic structure contains prominent and distinctive audible, ultrasonic, and non-linear elements.  Journal of the Acoustic Society of America 128: 1.3455855.

Photo of cactus deermouse at top by J.N. Stuart, via Creative Commons 2.0 license (Flickr).

This post may contain affiliate links. If you purchase a product through an affiliate link, your cost will be the same, but Wild Mountain Echoes will automatically receive a small commission. Your support is greatly appreciated!

Translate »