Why did roe deer not migrate into North America?

As I have mentioned before in some of my posts, I am highly curious about the ancient migration of species.

As I have been reading on the movements between North America and Eurasia through Beringia by many mammalian species including large ungulates such as ancestors of moose (Alces alces) and red deer (Cervus elaphus) and elk (Cervus canadensis), I wonder why roe deer (Capreolus spp.) never ended up in North America.

At least I am not aware of such geohistorical distribution although it is possible that I have not been appraised on the descendance of some North American cervid species respective to Capreolinae subfamily to which, e.g., Odocoileini tribe belongs that represents many North American cervids.

However, morphologically speaking, I cannot truly perceive common ancestry among these species.

There might be four explanations:

1.

Roe deer are forest dwellers while Beringia was covered with tundra-steppe vegetation.

However, European roe deer (Capreolus capreolus) are a stricter forest species.

Siberian roe deer (that currently reside closer to the ancient migration hotspot) are a steppe (and forest) species adapted to consuming grasses and forbs as well as capable of enduring severe weather (which is also true of the European roe deer).

2.

Roe deer dispersal is too limited to have reached North America.

It is true that roe deer are not the dispersal champions.

I have not yet read too much about the Siberian roe deer but the European roe deer frequently do not disperse further than 1 – 1.5 km from their natal area.

At such slow dispersal rates, it is understandable how they would not have reached other continents.

However, there might be phenotypes with different dispersal capacities in the European roe deer species (Wahlström, L.K. & Liberg, O., 1995) and I have yet to learn about dispersal rates in Siberian roe deer which cannot be lower than those in the philopatric European roe deer.

The ‘dispersing phenotype’ can cover average distances up to 120 km per year and in the Scandinavian population discussed in the publication referenced above, mean dispersal for the more sedentary population was also ca. 4 km per year.

Other publication by the same authors (southern Sweden study area) gives even lower average dispersal distance (ca. 2 km; Wahlström, L.K. & Liberg, O., 1995).

For example, red deer (Cervus elaphus) are philopatric, as well, more so in the female social group.

Dispersal rates in red deer definitely do not exceed 120 km per year and while it is difficult to find precise data, they might vary around 15 – 30 km in males.

Thus, if the ancestors of red deer ended up in North America assuming relatively short dispersal distances, so could have the ancestors of the roe deer (if accounting for a possibility of a dispersing phenotype).

3.

Competition and predation could have determined dispersal limitations in a rather small ungulate that would be outcompeted by most other cervids.

Predation might not have been that different between Asia and North America during Pleistocene.

However, the competition among cervids might have been a contributing factor which kept the roe deer Eurasia-bound.

4.

Relief is a significant obstacle to dispersal in many species.

This is a map of Beringia 18 000 calendar years BP to modern times.

In fact, the pre-historic map is not even needed because the contemporary terrestrial connections to Bering Strait are already rather informative.

However, roe deer also inhabit more mountainous regions (e.g., Carpathians, mountainous areas in the Mediterranean, Central Asia).

Of course, the situation during glacials would have been very different than it is today regarding both highlands and lowlands.

On mountainous regions roe deer still keep to forest areas (below timberline) and crossing the mountains could be difficult for a small species, not explicitly mountain-adapted (unlike goat and sheep species) that relies on tree and other vegetation cover (even Siberian roe deer that are deemed a steppe species, seek forest cover in modern day).

This could have been a contributing factor.