I don't think that enough matings have been done to prove anything, yet.
Here is a parallel situation:
A retired pro geneticist that I know raises zebra finches. He is trying to determine whether the crested mutant gene, which is some sort of dominant mutant in zebra finches, is lethal when homozygous. For what it's worth, there is a crested mutant gene in ducks and another in canaries. Both of those mutants are lethal when homozygous. Here is how my friend is doing it.
1. Mate crested x crested. The crested offspring have a 33% probability of being homozygous. He keeps these for breeding tests.
2. Test the possible homozygous crested finches by mating them to normals (non-crested). Seven crested babies and zero noncrested babies produces a 99% probability that the crested parent is homozygous. If a mating produces one or more non-crested babies, then the crested parent must be heterozygous (with a crested mutant gene paired with a normal gene). Failed matings are replaced with new test matings.
3. Continue the test matings until either there are some >99% probability homozygous crested finches or there is a less than 1% probability that he missed getting a homozygous crested finch simply by the luck of the draw. So far, there have been nine test matings. In each, the possible homozygous crested parent was not homozygous. My friend has been at this for several years and is shooting for 20 failed test matings before deciding the crested mutant gene is lethal when homozygous.
I would use the same protocol for testing whether or not the spider mutant gene is lethal when homozygous. This is a very time consuming procedure, so it's fairly obvious why nobody has done it yet. 
Paul Hollander