The reference written by Ugo Croce from his website contains several inaccuracies that should be mentioned here for the purpose of clarity.

1) Albas are hexaploid, yes, but they have four pairs of chromosomes from Canina and two from Gallica, not the other way around as written.

2) Hybrids certainly can have more chromosomes than either parent, and interestingly, this is what happens in Albas because of the unbalanced meiosis in Canina-group species. These species are pentaploid, and ovules are tetraploid (four pairs), while pollen is haploid (one pair). If a tetraploid Gallica or Damask was the father, and a pentaploid R. canina was the mother, we'd have a hexaploid seedling. The Gallica or Damask pollen would be diploid (two pairs of chromosomes), while the Canina ovule would be tetraploid (four pairs of chromosomes). So the hybrid would have more chromosomes than either parent. Because virtually all existing Albas present as hexaploid, this is the presumed origin of the class -- Canina mom, Gallica or Damask dad. And there are numerous instances of "un-reduced gametes" in garden hybrids resulting in seedlings with more chromosomes than either parent -- certainly not the norm, but also not unknown.

3) Albas certainly show more Canina characteristics, but that's to be expected, since they are 2/3 Canina. There are some Damask traits to be found, if you look closely. For one, R. canina hips and pedicels are generally smooth, whereas most Alba hips and pedicels are bristly. If you compare the hips and pedicels of R. canina, 'Alba Semi-plena', and 'Kazanlik', you'll see the influence of the third on the second, while certainly maintaining the size and shape of the first.

4) While there certainly are diploid species from warm climates and tetraploid, pentaploid, hexaploid, and (I think) octoploid species from cold climates, it is not necessarily the case that higher ploidy always equals more cold hardiness. Some very cold hardy species used in breeding very cold hardy garden hybrids are diploid, such as R. rugosa, R. blanda, and R. setigera. What's more likely is that higher ploidy resulted from the uncommon "un-reduced gamete" phenomenon, or chromosome doubling following hybridization between two different species, as has happened with 'R. x kordesii', for example.

:-)

~Christopher