Signatures in snow

14 March 2014 by Malcolm Campbell, posted in Biology

Perhaps I write for no one. Perhaps for the same person children are writing for when they scrawl their names in the snow.” From The Blind Assassin by Margaret Atwood (1939- )

A late winter snow has stretched a fresh canvas on the landscape. Ivory white. Pristine.

Like a blank canvas pulled across a frame, the snowfall begs for artistry.

As children, newly fallen snow invited etchings. Sticks delicately dragged across the powdery surface engraved a transient panorama. Or portrait. Or autograph.

With a wind, they were wisped away.

Those desiring a more durable mark wrote their names in pee. Always kids with conveniently short names – “Bill” or “Bob”. Never “Nicholas” or “Alistair”. Their yellow signatures proclaimed their boldness. A claim to audacity for all of us to see. yellowsnow1

With the recent snow, the dogs leave similarly audacious marks of a less artistic design. And yet they each have their own individual style. They are no less signatures than the human-derived variety.

Normally dog urination is invisible to us. Left in place after a swift squat or lift of the hindleg, it is virtually imperceptible to the human eye – maybe just the sheen of a damp spot where the urine has been deposited. In the snow, it is a different matter altogether. The snow records a recent history of dog urination. It provides our species, reliant on visual evidence, with insights into the invisible world shaped by urine that so many of our fellow mammals negotiate every minute of every day.

The primary function of urination is to expel waste. During normal physiological functioning, in mammals like humans and dogs, the kidneys function to maintain the steady state – homeostasis. Amongst the important things that kidneys do is to remove soluble waste products from the blood. This includes excess water, and nitrogenous wastes, including urea and uric acid, from the breakdown of proteins and nucleic acids during normal cellular functioning. The waste collected by the kidney travels via the ureters, tubes of smooth muscle fibres, which propel the fluid into the urinary bladder. The bladder is a highly elastic organ, designed to hold liquid – urine.

As urine fills the bladder, pressure builds on the wall of the organ. This pressure is sensed by both the sympathetic nervous system and the parasympathetic nervous system. The sympathetic nervous system informs the brain that it’s time to urinate, and the brain, in turn, exerts control to hold off or to proceed. Of course, the conscious desire to urinate can be ignored, or it can be a powerfully urgent one, depending on the extent to which pressure has built in the bladder. The parasympathetic nervous system relays a signal to the smooth muscle that surrounds the bladder to induce the muscular action that will void the bladder. yellowsnow2

When the signals to void have been provided, the bladder passes urine through the urethra, the route to exit from the body. In male mammals, urine is ejected through the penis. In most mammal females, with some exceptions, urination occurs at the vulva.

Urination is the means by which mammals expel soluble waste. Most mammals have co-opted this process as a means by which to communicate both within and between species.

Urine is loaded with useful information for communication between animals.

Colour is perhaps the least informative aspect of urine. Urine’s characteristic yellow colour is derived from urobilin, a breakdown product of the haem in haemoglobin that is filtered from the blood by the kidneys. The intensity of yellow reflects the extent of dilution of the urobilin. When the individual is well (or over) hydrated, the urine is pale. When the individual is less hydrated, or dehydrated, the urine can be intensely yellow.

The real information in urine comes from what cannot be seen.

Urine is loaded with chemicals that supply a detailed profile of its creator. During its production in the kidneys, and as it passes from the bladder to the urethra, urine accumulates organic molecules that are characteristic of that individual. Some of these are derived directly from the individual’s own metabolism. Others molecules are generated by bacteria that reside in the urethra, which are acquired as the urine flows through.

Proteins and hormones in urine create a chemical signature that tells of the organism’s species, its sex, its health, its fertility, its social status, and the extent it is related to those that may encounter it.  This chemical signature is largely scent based. Not surprisingly, a good number of mammals have co-opted this signature as a means by which to communicate. yellowsnow3

For many of our fellow mammals, urine signals function to set up social relationships. For example, African wild dogs use urine signals to indicate social status within groups. African wild dogs discriminate between urine signals that are merely released on the bare ground, which may simply be for urination purposes, and those that are deposited on vertical plant matter, which are more likely to be a signal. In keeping with this, they are more attentive to urine on vegetation than on the ground.  What’s more, they are most attentive to urine left by dominant females. Even more intriguing, pair-bonded, mating individuals are likely to urinate in tandem, one after the other. The order in which the urination occurs is contingent on the relatedness of the pairs. In unrelated pairs, males tend to “overmark” the female – depositing his urine on top of hers. Intriguingly, this is reversed in related pairs – the female overmarks the male, and males are less likely to overmark her urine.

Similar urine conventions are in play with other canines. Wolves gather, and leave, information about relatedness, fertility and social hierarchy through urine signals. While the data is surprisingly scant, even domesticated dogs follow some of these conventions.

In wonderful research reported in her PhD thesis, Daisy Berthoud examined urine signal conventions in domesticated dogs. Dr. Berthoud conducted her “field studies” in two locations – a dog park in Cambridge in the UK, and another in Paris, France. With painstaking observation, she documented hundreds of cases of dogs’ visits to the parks and recorded where and when they urinated.

In general, Daisy Berthoud found that male dogs were most likely to engage in both the deposition and interrogation of urine markings. While female dogs would sniff a urine mark, they wouldn’t necessarily deposit one themselves. Male dogs, on the other hand, frequently deposited a mark when they smelled a recent make made by another dog. Like African wild dogs, domesticated dogs were most lkely to deposit a urine mark on an object (like vegetation) than on the bare ground. yellowsnow4

Perhaps the most fascinating aspect of Dr. Berthoud’s had to do with overmarking. Male dogs overmarked fresh urine marks of other dogs, but not their own. This is in keeping with the notion that the urine mark may function as an indicator of territory, a means to indicate that “I see your mark, and I am leaving my own to hold sway in this territory as well”. Also in keeping with this, dogs would overmark their own urine mark if it was older – a day or two later, for example. In the territorial urination model, this is a means by which to reaffirm that “the territory is mine”.

This said, as Dr. Berthoud notes, there is no evidence to suggest that such marks are truly "territorial". That is, it may be that urination does not function in a manner that truly defines territory - at least not in a defensive sense. Such marks may simply convey another sort of social information - more "I was here", than "get out of here". Either way, it would appear that dogs have a sense of where to place their urine.

What is striking is that these findings seems to suggest that dogs have a notion of “self” versus “other” based on urine cues.

Canine behaviourist, Marc Bekoff, has long been interested in the manner in which dogs monitor themselves – the extent to which they have some form of self-realisation. Focusing on urine left by his own dog, Dr. Bekoff took advantage of the opportunity for humans to easily identifyt urine as “yellow snow” afforded by winter snowfall. By collecting and moving urine left in snow banks by his dog and those of others, Dr. Bekoff was able to manipulate the urine cues his dog was exposed to. Over the course of five winters, Bekoff moved urine marks around and observed how his dog reacted to them.

Now, as Marc Bekoff notes, the findings are based only on a sample size of one, and should be regarded as anecdotal. Nevertheless, what Dr. Bekoff found is intriguing – his findings aligned with Dr. Berthoud’s larger field research beautifully. That is, Dr. Bekoff’s dog was much more likely to overmark the urine of other dogs, as opposed to his own urine. What’s more, he was more likely to do this if the other dog was a male, the same sex as Dr. Bekoff’s dog.

Obviously, such a study is in need of significant replication. The research could also benefit other improvements in experimental design – such as dramatically increasing the sample size, and separating the tasks of experimental manipulator and observer – so as to remove any biases. Importantly, as Bekoff notes, these are not insurmountable problems, and the simple experimental design he used could easily be replicated by many, many people – in a large citizen-science sort of project. Indeed, many people have probably already made similar observations of their dogs urination and overmarking habits, albeit without the experimental manipulation of where the marks are located. yellowsnow6

The findings do point to a form of urine-based self-recognition in dogs.

Is it possible that dogs can identify self, or minimal things that are familiar, simply via scent, just in the way that we recognise such things in a mirror?

Recently, Greg Berns and colleagues investigated the canine capacity to discriminate between themselves and others. They did so by scanning dog brains using functional magnetic resonance imaging (fMRI).

fMRI is a technology that enables us to visualise brain activity. fMRI is based on the simple observation that those regions of the brain that are most active tend to have the greatest flow of blood, and that the blood that flows to these regions tends to be more oxygenated. fMRI  works by using differences in the magnetic properties of haemoglobin when it is associated with oxygen. When haemoglobin is oxygenated it is less resistant to magnetic fields, and interferes very little with magnetic resonance signals. Consequently, when oxygenated blood flows to a region of the brain, oxygenated haemoglobin’s lower interference with magnetic resonance results in a greater signal being observed in that region. By comparing magnetic resonance signals measured in three dimensions, throughout the brain, fMRI can reveal those regions that have the highest amount of oxygenated haemoglobin, and, are therefore, presumably, the most active.

Greg Berns and colleagues were interested in those regions of the brain that might be responsible for discerning familiarity on the basis of odour. To do so, they trained a dozen dogs to sit calmly in an MRI. They then exposed each of the dogs to five different scents: 1) the dog’s own scent; 2) a familiar dog’s scent; 3) an unfamiliar dog’s scent; 4) a familiar human’s scent; and 5) a unfamiliar human’s scent.

Following exposure to the scent, the dogs were subjected to two kinds of. The first fMRI analysis was focused on two regions of the brain – the olfactory bulb and the caudate nucleus. This targeted analysis enables better resolved, more robust data on changes in brain activity in that region. yellowsnow7

The second analysis was a whole brain scan. While a whole brain fMRI scan enables a more comprehensive examination of the impact of the scents across the brain, it comes at a cost – the data are not as robust statistically. Nevertheless, it can both be confirmatory of targeted fMRI, and reveal regions of the brain that might also be involved in the response.

The fMRI analysis that focused on specific regions of the brain found that, regardless of the scent used, all scents created greater activity in the olfactory bulb.  And a good thing too, asthis is to be expected. The olfactory bulb is a region in the forebrain of vertebrates that is involved in the detection of odours. The fact that all five scents induced a response in this region was a good positive control, indicate that the dogs could, in fact, perceive the scent.

Intriguingly, the scent of familiar humans strongly activated the caudate nucleus. In fact, the scent of the familiar human had the greatest effect on the caudate nucleus relative to the other scents. This is fascinating, as the caudate nucleus has been implicated in shaping positive expectations, including those related to social rewards. Greg Berns and his colleagues note that caution must be applied in attributing this activity to a positive emotional state in response to a familiar human’s scent. Instead, they suggest that the activity may be related to a “seeking” state – a motivational driver that is derived from the scent of a familiar human.

Whole brain scans confirmed the results that focused on the olfactory bulb and the caudate nucleus. That said, they also revealed something remarkable about a dog’s subjective experience – the experience of being a dog. That is, the whole brain scan found that the medial frontal cortex was most responsive to familiar dog scents, including self. This region of the brain is associated with cognitive control in primates. It is unknown what role it plays in dogs, but it may be involved in forming some cognitive basis of self-recognition. Minimally, it seems to be involved in some activity related to the perception of “us” versus “them”, when “us” is a dog. Future studies will undoubtedly further refine this potential role. yellowsnow9

For the time being, we can revel in the fact that other mammals have a sense of self and other that is shaped by something as seemingly banal as a waste product – urine. Perhaps what is most remarkable about these considerations is that we humans pass by untold multitudes of urine-derived signals every day without paying them any heed. It’s remarkable that we are so oblivious to them, as the lives of our fellow mammals are shaped, and undoubtedly, enriched, by these signals on a constant basis. They determine who interacts with who, who defers to who, who mates with who. Like ripples in ponds, they create currents that move things. They provide a layer of complexity, a depth, to existence that is profound.

Urine-stained snow provides us with an opportunity to see a marvel, which, literally, passes under our noses undetected, all the time. It reminds us that sometimes things that appear as waste can still retain great value when used for another purpose. Urine may be a waste product, but evolution has ensured that urine’s value as a powerful conveyer of information has not been wasted at all.

The old adage advises us to “not eat the yellow snow”. This is certainly sound advice. This said, while you may want to avoid eating it, there is still great value in stopping, looking at it, and pondering over its deeper meaning. It is a dog’s name for himself, written in snow.

Images: All photographs by Malcolm M. Campbell. yellowsnow8


Bekoff M (2001) Observations of scent-marking and discriminating self from others by a domestic dog (Canis familiaris): tales of displaced yellow snow. Behavioural Processes 55: 75-79

Berns, G. S., Brooks, A. M., & Spivak, M (2014) Scent of the Familiar: An fMRI Study of Canine Brain Responses to Familiar and Unfamiliar Human and Dog Odors. Behavioural Processes. doi: j.beproc.2014.02.011

Berthoud D (2010) Communication through scents: Environmental factors affecting the urine marking behaviour of the domestic dog, Canis familiaris, kept as a pet. PhD Thesis. Anglia Ruskin University.

Jordan, N. R., Apps, P. J., Golabek, K. A., & McNutt, J. W (2014) Top marks from top dogs: tandem marking and pair bond advertisement in African wild dogs. Animal Behaviour 88: 211-217

Jordan, N. R., Golabek, K. A., Apps, P. J., Gilfillan, G. D., & McNutt, J. W (2013) Scent‐Mark Identification and Scent‐Marking Behaviour in African Wild Dogs (Lycaon pictus). Ethology 119: 644-652


One Response to “Signatures in snow”

  1. Malcolm Campbell Reply | Permalink

    Dr. Daisy Berthoud, whose excellent thesis work is cited in this post, has sent a super email to help clarify some of the elements of this piece - some of which were potentially misleading. In particular, the notion that domesticated dog urine marks are "territorial" is not necessarily supported by Dr. Berthoud's research. As she notes:

    "Yes, dogs did overmark fresh urine marks more often than older ones, however, there was no other evidence to show that those pet dogs marked in a territorial manner while out on walks. This may simply occur because they are the most easy to detect compared to urine deposited more than a day ago."


    "It may be that you interpreted my findings (male dogs overmark fresh urine of other dogs, but not their own) in a way that lead you to conclude that dogs may deposit urine in order to demark their territory. In contrary, one of the main points in my thesis was to review the well-established notion that dog urine marks serve a territorial function in pet dogs."


    "It is unknown how long urine marks can be detected by conspecific dogs; therefore, the frequency with which urine marks may need to be refreshed to be effective signals is also unknown and it may not be possible to conclude that dogs mark in a territorial manner."

    In keeping with this, I have added a paragraph that tempers the notion that domesticated dog urination is "territorial", focusing instead on its capacity to function as a social communication mechanism. The point remains that dogs sniff urine and deposit urine in specific locations - indicating some notion of "self" versus "other". Consequently, the central thrust of the piece remains the same.

    As an aside, the idea that a cue that is used to mark territory in some canine species has been co-opted for a different social communication purpose in domesticated dogs is an intriguing one - worthy of further consideration. Perhaps the topic for a future post!

    In any case, much thanks to Dr. Berthoud for the clarification. I will leave the final words to her:

    "I have been very surprised (in a good way) and happy to discover that you have cited my Ph.D. work in one of your recent articles...I have enjoyed reading your article very much and would like to thank you for reading and commenting on it."

Leave a Reply

four − = 2