Differences in the Behavior of Landraces and Breeds of Dogs - Genetics and the Behavior of Domestic Animals - page 197

dogs in Italy. In January 1989, in the mountain town of Mucuch´es,
Venezuela (human population, 2000), we estimated a population of 800 dogs
(1 dog/2.5 people). In 1994 it was estimated that 1,360,000 free-ranging
dogs lived in the communal lands of Zimbabwe ( Butler and Bingham, 2000 ).
Human population there was estimated at 6,190,000, or 22 dogs/100 people,
or 1 dog per 4.5 people. The current data on dog populations in developing
countries shows a range of 8.1 dogs/100 people to 35 dogs/100 people
( Jackman and Rowan, 2007 ). In resource-rich locations such as the Mexico
City dump, we estimated 700 dogs/square kilometer and that is by no means
a record.
The United Nations World Health Organization classifies dogs in four
categories ( WHO, 2004 ), and in the town of Mucuch ´ es all four were repre-
sented: feral dogs, which are independent and unrestricted and live outside
the village (although individuals often came in during twilight hours); neigh-
borhood dogs living in town and observable at any time of day independent
of humans even though they may recognize and beg from individuals; family
dogs that are more or less managed by humans but semi-restricted; and the
fourth category, the fully restricted dog that is totally dependent on humans.
This type of habitat partitioning by dogs is familiar (e.g., Boitani et al.,
1995; Macdonald and Carr, 1995; Macdonald et al., 2004 ).
People often refer to the semi- or unrestricted free-ranging village dogs
as strays, mongrels, pavement specials, or curs. The assumption is that these
are dogs which are the results of irresponsible ownership, or have severed
their relationship with humans, and live by foraging on garbage within or on
the margins of the village. But worldwide these dogs are phenotypically
remarkably similar in shape and size, suggesting to us that they are the result
of natural selection and adaptation to a niche. The uniformity of size, shape,
and behavior also suggests a lack of artificial selection. There is historical
evidence that smallish canids have continuously occupied villages since the
beginning of writing ( Coppinger and Coppinger, 2001 ).
Researchers have studied various ecotypes and varieties of village dogs,
using names that reflect their assumed ontogenetic or phylogenetic history,
or their foraging strategy: they are stray dogs, pariahs, sylvatic, feral, or wild
dogs. Within the village environment are probably several ecozones which
elicit different foraging strategies, and over time and geography, differences
in the riches of the niches account for a variety of the observed adaptive
strategies. Ecologists interested in the behavior of carnivores make predic-
tions about these social arrangements based on the “resource dispersion
hypothesis” ( Carr and Macdonald, 1986; Macdonald et al., 2004 ).
Three of our study sites are Zanzibar ( Coppinger and Coppinger, 2001 ),
Ethiopia ( Ortolani et al., 2009 ), and South Africa ( Gallant, 2002 ). The dogs
of Zanzibar are presented here as the type form of village dog. The study
site in Zanzibar, in 1996, was primarily rural Pemba (island) where people
live in villages off the road or in small towns along connecting roads. The