I think the 3DS is starting to reach the point where the install base isn't really an obstacle for most games to reach their full potential. No one would have predicted vastly higher sales if the install base had been at a few million higher
And this is what makes me suspicious, a smaller install base should result in lower sales, at least, if we are talking about 5 million 3DS against almost 30 million DS (September 2009).
The theory would be, that the crazy software numbers are part due to the abnormal sales development of the 3DS.
Usually, there isn't such a sales spike, shortly before the 2nd generation of games arrive on a console. The 2nd generation of games is often much more polished, then the launch titles and it is often the generation, where companies start to release their bigger franchises (Monster Hunter, Resident Evil, Love Plus). The companies often aren't willing to throw these titles on the market at launch, because the user base is much to small at this point to justify the high development costs for such franchises, so they wait.
The 3DS is in the lucky position, that this generation arrives, shortly after the system really exploded sales wise. People that just bought a 3DS are seeing a lot of big names released in a short time and they are just more willing to buy a lot of these games, because they are looking for new games to try out their new console anyway.
I don't know, if this is the case, but as much, as I am happy, that at least one system seems to be able to move software in Japan, I think the numbers are driven
partly due to the performance of the 3DS in the past months, which I think will normalize as software sales will be in the coming months.
I would love to be proven wrong here.