If there is one thing to understand about high volume training its mitochondria mitochondria mitochondria. I could list countless articles showing mitochondrial density balloon with high volume training.
Know this though--mitochondrial biogenesis stimulated at low exercise intensities. Its primarily a question of training volume. The pace of the run does effect where those mitochondria are distributed. . . whether they develop in type I or type II fibers, commonly known as slow and fast twitch muscles. The so-called scientists that knock LSD use some rat studies and say that at slow paces you don't recruit fast twitch fibers and hence that don't develop a high oxidative capacity and thus can't be called upon in racing. This is very bad science.
First you need to understand some key differences between rats and humans (I am predominantly a rat researcher so this isn't biased). Human muscles are heterogenous for type I and type II fibers. For example, the human vastus lateralus or gastroc or etc. has a nearly equal amount of type I and II fibers in the average person. The human soleus does have more type I than type two fibers. Rats are different. A single rat muscle is generally homogenous for a either type I or type II fibers.
So when some scientists run some rats at different speeds and say that the oxidative capacity of the type II fibers wasn't developed at 70% of VO2max or whatever, it doesn't fly for humans. At certain speeds for rats certain muscles bare more of the load than others so of course the type II fibers aren't going to be challenged as much.
Things are MUCH different for humans: (If your getting sick of reading by now you need to read this!) Recruitment of different fiber types in the HUMAN vastus lateralus, a major quad muscle, was examined at different percentages of VO2max. The results will probably surprise you.
At 43% of VO2max ALL type I fibers were being recruited and 20% of type IIa (fast-oxidative) were being recruited.
At 75% of VO2max ALL type I and ALL type IIa fibers were being recruited. This represented 85% of the entire muscle mass.
The non-oxidative fast comprised the remaining 15% of the motor unit pool.
(Maximal aerobic power: neuromuscular and metabolic considerations. HJ Green and AE Patla. Med. Sci. Sports Exerc. Vol 24. No. 1, pp 38-46. 1992)
Additionally it turns out that after 90 minutes of moderate intensity exercise that these weakly oxidative fibers begin to be recruited. This is why 120 minute runs are great. Don't stop at 90 though or you won't see the benefit. Several studies have shown that the process just begins at 90 minutes.
Also consider that well trained runners are recruiting 90%+ of their muscles mass at 75% VO2max because their pace is faster and requires for force.
So, the point here is that in HUMANS running at slow to moderate paces does indeed involve a large amount of fast twitch muscle. Its all recruited when you start running longe then 90 minutes. Thus, mitochondrial biogenesis is induced throughout the full spectrum of fiber types. With time the type IIb are even converted into type IIa. Salazar was biopsied once and found to have NO type IIb fibers (this is a result of high mileage over years).
Running faster does induce a different set of adaptations that are necessary for optimal performance. But the foundation of it all (on the peripheral side) is mitochondrial density.
Mitochondrial benefits come very quickly, its just a matter of where the whole body is prepared to handle the increased trianing load. Say a guy name JOE is running 50 miles per week. He does this for twelve straight weeks. His increase in mitochrondria between weeks 11 and 12 is minimal if there is any at all. Well, Joe for some crazy reason runs 100 miles a week for the next two weeks--the increase in mitochondria is measureable and quite large. It will just take a while for his legs to recover so that he can experience the benefit. So, build up slow, but certainly build up. Only worry about the pace after you have topped off volume for quite a while.