While considered today a leisure and sport activity, Cross-country skiing (CCS) represents an interesting way of locomotion both to the biomechanist and to the exercise physiologist. The former has the opportunity to apply the body of knowledge acquired for walking and running in terms of mechanical paradigm, energy exchange, etc. to the explanation of the high speed performance attained in CCS and the significance of the different techniques. Differently, the latter is mostly focusing on the metabolic energy cost associated to the different gradients and speeds and is interested to identify and interpret the strategies adopted during long-lasting CCS competitions. While human body certainly did not optimally evolve to move sliding on the snow, it has been reported that CCS has been used for many thousands of years from the nordic ethnic groups as a mean of transportation. For this reason, before considering it just a leisure and sport activity, CCS should be regarded as an adaptation of walking and running to a problematic environment, and investigated as an additional gait. While in the literature some works appeared about the metabolic cost of CCS, almost nothing has been published on the mechanical work associated to it. Mainly for technical reasons, the same methodologies used to evidentiate the mechanical determinants of walking and running energetics have not been successfully emploied in CCS research. The main difficulty is in obtaining the 3D coordinates of the athlete body segments with a reasonable spatial and temporal resolution, spanning at least a complete stride in the natural environment. As a matter of fact, a single locomotor act can span more than 8-9 metres and many cameras should be needed to capture it fully. In addition the reflection of the ambient light on the snow crystals prevents the use of commonly adopted motion analyser, as the ELITE System (BTS, Italy). However, some measurements have been done by using analogic video cameras, whose frames, captured at 25-30 Hz, have been successively digitized by hand. The obtained 2D data have been postprocessed as to obtain the 3D coordinates of body segments, the essential step needed to assess the mechanical work of locomotion. Despite of a lower than usual spatial and temporal resolution of the system, some important insights in CCS mechanics are provided by such a technique. A couple of kick-double pole strides on a level surface have been video captured during the World CCS Championship in 1996. The single movie frames have been processed as described previously. The obtained 3D coordinates of the segments have been filtered and the coordinates of the body centre of mass have been obtained. Then a program designed in our lab calculated the associated potential energy (PE, due to its vertical position) and the kinetic energy (KEx, due to the square of its speed). By summing the two energy time courses and adding the increases in the resulting curve (named total energy of the body centre of mass, TE) an estimate of the mechanical external work was computed. The mechanical internal work, necessary to move the limbs with respect to the body centre of mass was obtained from the relative speed of body segments. Just below the animation regarding Manuela Di Centa stick diagram, the time course is represented. It is worth observing how the raising and lowering of the trunk is associated to the PE curve, while the vaious peaks in KEx are associated to different phases in the stride cycle: 1) the kick push, 2) the impact of the poles, 3) the combined push made by the flexion of both trunk and upper limbs, 4) the final push of just upper limbs. This kind of analysis suggests that important clues on the efficacy of the motor act along the stride cycle can be obtained from the inspection of the mechanical energy time-course. The mechanical internal and external work of kick-double pole, labelled CCS in the graphs, are lower and higher, respectively, than in level running (labbelled R). The first effect is due to the lower stride frequency of CCS, while the second one is caused by the unbalance between the single peak curve of potential energy and the 4-peak shaped kinetic energy curve. However, differently from running, some exchange between PE and KEx occurs, reflecting some mechanical energy saving mechanism (as witnessed by the higher Energy recovery, a parameter estimating the pendulum-like potential-kinetic energy interchange). In the table below the internal work has been partitioned in single limb contribution. It emerges that the pushing limb, during the kick, account for about 1/3 of the total internal work and does 4 times the work of the contralateral limb. Differently, upper limbs equally contributes to the mechanical internal work. Future perspectives In order to use such an analysis to evaluate athlete performance on an individual basis, the techniques needs refinements and more data about the different CCS gaits have to be collected. The spatial and temporal resolution of the acquisition system has to be increased. Then a consistent database of different CCS techniques, performed by a great number of subjects including amateurs and medium level athletes, have to be built. Ideally, the database should include measurements made at different speeds, gradients and snow temperature (studies on the dependence of snow friction on ambient temperature would be welcome). Simultaneous metabolic measurements both at controlled speeds and during long-lasting CCS competitions will contribute to get insights in the economy and efficiency of a n ancient and modern gait.