The terms "wandering dog" are often wrongly used for a dog which is not under its owner’s responsibility anymore; in national law, it is in "straying state".

A wandering dog is a wild dog with no owner. In France, there is no wandering dog.

Stray dogs are responsible each year of significant predations on domestic herds, especially on sheep. The data put forward in literature are quite variable: depending on the studies, they could be responsible of annual predations up to 0.18% to 5% (!!!) of the present number of animals. In other words, for a number of 10,000 ewes, straying dogs could hunt each year 18 to 500 ewes, according to the studies taken into account. These data must be used cautiously as the areas where the studies were carried out are not necessarily very representative.
The latest study on the topic was carried out in partnership with the CERPAM (pastoral department, Provence-Alpes-Côte-d’Azur), the SUAMME [1] and the ENITAC [2]; considering its scope and the diversity of the prospected areas, it is presumably the study that can be extrapolated most easily. Two numbers are to be remembered:
 annual average predation rate of 0.25% (25 ewes hunted each year out of a number of 10,000);
 number of attack per herd and per year of 0.18.

These data allow to tone down the debates on the respective responsibilities of dogs and wolves concerning the damages on domestic herds. In 2006, in the Alps, 500,000 sheep belonging to 1,820 breeders were situated in areas where the presence of wolves was proven. Before the arrival of wolves, the order of magnitude of annual predation by straying dogs could be of 1,250 ewes in this area. This number has to be compared with the 2,453 indemnified sheep for the predation by wolves in 2006.

These last data nevertheless concerns herds that are by a large majority fitted with protection means whereas the level of predation due to dogs was measured in areas without wolves, where the herds very seldom mobilize specific protection means.

The number of attacks attributed to wolves per livestock and per year was of 0.50 in 2006; a breeder in an area where wolves are present thus suffers from a wolf attack every two years in average.

So it seems that the attacks by wolves could be more common than the attacks by stray dogs while the involved herds are more often protected; on the other hand, the number of victims per attack is lower for wolves (2.7) than for dogs (3.8).

A recent study was carried inthe Basque Contry in Spain gives a method to discriminate wolves and straying dogs and their diets by faeces’ analysis.

Firstly, an genetic analysis of the faeces ensures the species differentiation. Then the analysis of the rests in faeces ensures the study of the diet and espacially the domestic herds’ part. On 136 faeces analyzed, 31 belonged to wolves and 53 to dogs (and 2 to foxes).

In the lupine faeces, about one quarter contained domestic herds’ rests against a few more than an half for canines faeces. In proportion straying dogs seem to consume more domestic herds than wolves; that can be explained by the proximity between straying dogs and domestic herds.

The domestic ongulates’ proportion in straying dogs’ faeces is still high, which brings to light a significant impact on the wild fauna.

Beyond the datas, the interest of this study is to give a method which could ensure to study the impact of straying dogs on the domestic herds.

Beyond the differences in the ways of attacking between dogs and wolves, another reason is the implementation of protective measures in areas colonized by wolves, the effect of which is to reduce the number of victims per attack.

It is likely that the implementation of protection dogs on areas not colonized by wolves would similarly allow to limit the impact of straying dog attacks but the level of predation generally observed does not urge breeders to equip themselves with often restrictive protection means.

These data do not reflect the difference in the perception of attacks by dogs and wolves by breeders.

Simplifying the situation:
 dog attacks take place by day;
 the dog is seen during the attack;
 it is often question of a dog of the neighbourhood.

The problem can therefore be settled and it is possible to protect oneself against these attacks without altering the pastoral practices.

As for the wolf attacks, still simplifying :
 the attacks take place by night ;
 the breeder does not see the wolf.

To protect oneself from these attacks often requires a modification of the way to lead the herd and an important extra work, which is not necessarily enough to settle the problem.

- Study by the CERPAM: Dog attacks on flocks in Luberon and comparison with the predation on wolf territories, L. Garde - 2005 (in French)
- Minutes of the seminar "Wolf-breeding: open up to complexity", p.30 to 40: The damages made by straying dogs : results of investigations on 6 territories
- Noninvasive monitoring of wolves at the edge of their distribution and the cost of their conservation: spanish study on the in situ differentiation with the génétics of the domestic herds’ proportion in the diet of the wolves and the straying digs (in english) - 2009