There is an interesting new paper on a subject that often drives much discussion in this blog at least: how bad is inbreeding. The answer of course depends on several issues such as the intensity of inbreeding (not a mere true/false dichotomy) and the environmental pressure.
This paper deals with these matters in an experimental set with Americamysis bahia, a species of mysid shrimps.
Jeffrey E. Markert et al.,Population genetic diversity and fitness in multiple environments. BMC Evolutionary Biology, 2010. Open access.
The authors created a number of extremely inbred populations by
1. Taking a single pregnant female (presumably impregnated by a single male, as is typical in this species) from the main population (AMX) and removing it when the brood was released
2. Taking then one impregnated female from one of the resulting populations and another from a different one and placing them in the same tank until the brood was released.
3. Taking then one single impregnated female from the population resulting from step 2.
Overall it means that the hyper-inbred brood at the start of the experiment, described as 1X, had went through a narrow 2-4-2 bottleneck, being the equivalent to having 2.4 founders.
Populations of the type 2X, 6X and 8X were generated by mixing the respective number of 1X inbred populations, so they represent lesser degrees of inbreeding each.
By the end of the experiment, after 40 weeks, these were some of the results (more tables, graphs and explanations in the paper):


Figure 1 - Population fitness, estimated with Median Population Size (A), Last Census size (B), and Reproductive Index (C) [not shown here].
Paired box plots define the median and middle two quantiles in stressful (left) and permissive environments (right). Lower case letters unite groups that are not statistically distinguishable using post-hoc tests (Tukey’s HSD) at α = 0.05.
We can see that extremely inbred populations performed badly, specially in stressful environments (low salinity) but that not so extremely inbred ones performed reasonably well, specially in permissive (normal) environments.
While the results should not be strictly extrapolable to other species, we can reasonably conclude that a very low number of effective population founders (5 or so) seriously hampers survivability, specially in challenging environments, where essentially means a death sentence. However a not so tiny number of founders (25-30) can do quite well, in particular if the environment is favorable.
