Crossing
Nicotiana tabacum with
N. rustica, is a curious business. It's not simply crossing two differing strains of the same species, but attempting to cross two different species. Both are tetraploids, that is, they each contain four distinct versions of the chromosome sets inherited from their theoretical parent species. It's very messy.
Pairs of Nicotiana chromosomes, by species (haploid number).
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1086129/pdf/pnas01807-0011.pdf
When two varieties of the same
Nicotiana species are crossed, the tetraploid chromosome sets align properly. When attempting to cross these two different species, the chromosome sets do not align properly. As a result, the outcome of
N. tabacum ova pollinated by
N. rustica pollen is nearly always infertile--usually no seeds are formed, and the blossom pod dies and drops off the plant. The results are better, though still quite poor, when
N. tabacum pollen fertilizes
N. rustica ova.
So it's possible (though improbable) to fertilize a
N. rustica plant using pollen taken from
N. tabacum stamen.
If this improbable cross succeeds, then the resulting "mule" seedlings will consist of many, many distinctly different plants, each with a random mixture of the possible gene combinations. Many of these first generation plants will grow, but be infertile. The grower must select desirable plants from a large planting of this first (F1) generation, then attempt to germinate their seed. This process (of selecting plants from a large number of offspring of
each selected plant) must be repeated for a number of years, while the selection criteria remain unchanged. It is possible to eventually end up with low enough heterozygosity (mixed gene traits) to consider the newly developed plant to be genetically stable. Sometimes, back-crossing to one of the original parent plants is required to retain certain desired traits.
When crossing two established
N. tabacum varieties, each of which are genetically considered to be virtually homozygous (no mixed gene traits), the development process requires typically 7 to 10 years to complete. "Complete" means that all the seed of the self-fertilized, newly developed variety will grow true to type.
By contrast, an interspecies cross, if it works at all, will require at least 7 to 10 years, but probably much longer, as well as the substantial work of repeated, large plantings, to be fully stabilized.
This is not to disparage the attempt, but rather to point out (as others have already) that this is not simply a 1 or 2 season experiment with a small number of plants. Seeds from the first few generations of such a cross should be considered a pig in a poke (or as Forrest Gump would say, "like a box of chocolates. You never know what you'll get.")
Bob