I am 95% sure I am falling into a troll trap here but oh well. Since you refuse to go back one page here you go.
"A study by Riley et al., (2008) comparing HSR and TMR at the same speed (approximately 13.8 km/h) found that stride frequency was significantly higher and stride length shorter on a treadmill when compared to HSR. These results confirmed those of Schache et al., (2001) and Frishburg (1983) who likewise found increases in stride frequency and decreases in stride length when TMR.
Furthermore, both of these latter studies found a decreased swing phase and an increased stance phase when TMR versus HSR (Frishberg, 1983; Schache et al., 2001) indicating an increased period where the foot is in contact with the treadmill belt.
Biomechanical - Muscular Interactions and Differences:
As the tread belt rolls backward it drags the foot backwards, eliminating the need for the Gluteals and Hamstrings to pull the upper body forward thus making the movement easier; hence its potential lower energy cost (Frishberg, 1983). The Hip Flexors however, now have to work harder to bring the limb, which is being dragged backward, forward. Furthermore, the Frishburg (1983) study found that the lower limb was required to move through a greater range of motion in the stance phase when TMR (X=60.6 vs 54.5 deg). Similarly, Schache et al., (2001) found TMR to have a lower angle of hip flexion at initial contact and considered the increased hip extension position to be caused by the treadmill belt dragging the foot rear-ward. With this in mind, if the Hip Flexor muscles are tight, the increased hip range of motion requirement on a treadmill may cause the pelvis to rotate or the lumbar spine to hyperextend to release the tension on the hip as the limb is dragged backwards. This increased pelvic rotation and/or lumbar hyperextension, which typically occurs with fatigue (which increases stance phase duration) and at high speeds, places strain on the hips and lowerback and has the potential to cause a number of injuries.
It should be noted however that a study by Riley et al., (2008) found no significant differences in hip flexion/extension ranges or pelvic rotation between HSR and TMR. The subjects in this study (n=20) who were regular runners and ran at least 15 mi (24 km) per week, were, however, only assessed over a short period (no more than 15 consecutive gait cycles). Hence these runners would not have been subject to the impact of fatigue.
With the belt moving backwards, the ability of the gastroc-soleus complex (calf muscles) to push off is reduced (Riley et al., 2008; Baur, Hirschm, ller, M¸ller, Gollhofer, & Mayer, 2007) and these muscles must work harder in order to achieve an effective push off. This is one of the reasons why you tend to power walk when you transition from the treadmill back onto hard surface – your calf muscles have been activated more than usual for hard surface locomotion. Furthermore, research has shown that running on the flatter, predictable surface of a treadmill requires less ankle stabilization than that required for running across land, with muscles like the peroneals found to be less active when TMR (Baur et al., 2007)"