Expert discussions about the growing ecological and infrastructure problems caused by freight transport have led to proposals for three types of solutions: reduction of transport volume, a shift to more environmentally friendly transport modes, and application of the best available technology. In this paper, freight transport will be analyzed at its roots, i.e., firms as generators of transport activities, and to point out the mechanisms at work in this field. To provide a picture of freight transport, a survey of goods shipments and shippers in a defined region (East-Austria) was carved out. In addition to freight transport descriptors, such as volume, modal split, and firm-owned infrastructure (vehicles fleet, railroad siding tracks), the transport chain of individual shipments has been documented. Both the distribution of unimodal and multimodal transport chains and the number of transfer functions are presented. An evaluation of freight transport of firms was carried out on the basis of this survey, with the objective to find any regular patterns concerning firm structures and location patterns. The parameters used to evaluate freight transport are energy consumption and selected air pollutant emissions of individual transport modes. Using a single evaluation paradigm allows a comparison among different firms. The impact of freight transport is evaluated on the basis of energy use and transport-related pollution levels. The evaluation criteria were all employed on the basis of certain assumptions; this fact, together with the sampling approach to firms' shipments, created some methodological problems. Inaccuracies concerning the evaluation factors were offset by exact differentiation among the various transport modes. The extrapolation of shipments makes possible a comparison between the firms. Firms located in an urban agglomeration have a significantly higher transport-related consumption of primary energy than firms situated in peripheral areas. This difference also appears when firms are grouped according to dependence on location. Firm size plays a role only in outgoing transport, but specific primary energy consumption is unequally distributed among the various firm size classes, so that firm size as such does not permit conclusions about specific energy consumption. If the specific transport-related primary energy consumption of individual shipments is studied in relation to firm structures -- location-dependent, partly location-dependent and location-independent firm --, we find that primary energy consumption is lower for more location-dependent firms. Primary energy consumption in freight transport is essentially determined by mass and transport distance. Since specific mass varies within the NSTR commodity group, the commodity group is also a decisive factor in primary energy consumption. Firm sizes, type of industry, distance between a plant and the transport infrastructure, and vehicles fleet, however, have no or nearly no influence on primary energy use. In firms with siding tracks, the modal split share of rail transport is markedly higher than in others, while the share of road transport is correspondingly lower. The utilization of rail in freight transport depends on the existence of siding tracks. The modal split for all shipments received by the observed firms was 63.8 % by truck, 22.6 % rail, 13.3 % by ship, and 0.3 % by other transport modes; the modal split for shipments delivered was: 85.4 % by truck, 14.0 % by rail, 0.5 % by ship, and 0.1 % other modes. The modal split was calculated on the basis of the main transport mode on each route, i.e., the transport mode that was used on the longest segment of any given route. Rail is used more extensively in peripheral areas than in urban areas, and trains still play a larger role in peripheral (i.e., rural) areas, even though the absolute figures for rail use are low. Firms situated in peripheral areas have a significantly lower share of own account haulage than those in urban areas. On the other hand, loading capacities of company-owned vehicle fleets are the same in urban and peripheral areas. Urban agglomerations are characterized by the proximity of transport facilities: highway junction, train station, combined transport terminal and freight transport centers are all only a short way off. In contrast, peripheral and extremely peripheral locations mean that only the nearest train station is at an acceptable distance. This may be one of the reasons why rail is used to a greater extent there. Transport infrastructure to a certain extent controls the choice of transport modes.