This dissertation deals with market definition and demarcation, and discusses competition and market form classification in the spatial economy. The analytical tool for this study is an extended Location Theory model. A distinguishing characteristic of the spatial model of plant dispersion by market forces developed in the present thesis is the addition of the unit-value (e.g., the value per ton or cubic meter) of the commodity in question as a determinant of the long-run equilibrium solutions for a particular industry as to plant density, output volmne per plant and so on. Two ratios are the main determinants:

• The unit-value of the commodity (a) relative to the transport costs per unit-kilometer (c), alc
• The degree of economies-of-scale in production (b) relative to the density of demand (d), bld

In the model, the character of competition is determined not by the number of firms, but by the ratios alc and bld. These ratios are used in the first application of the model; a two-dimensional matrix in which the four market forms- perfect competition, monopoly, oligopoly and monopolistic competition - fit in as special cases. In addition to shedding new light on the traditional market form taxonomy; with the help of the model the following questions are discussed:

1. Which industries tend to cluster, and which indushies tend to be spatially dispersed
2. Why different market forms apply to different dispersed industries
3. How market forms may change over time

Empirical evidence has been gathered in Sweden and the U.S.A. The results give some support to the model's predictions as to the spatial structure of different industries.

The second application of the model is as a tool for Cost-Benefit Analysis of improvements in transport infrastructure so far as goods transport is concerned. The pioneering paper here is Mohring & Williamson (1969), in which the balancing of the nmnber of plants and the average transport distance by a multi-plant monopolist is mode led. Total benefits of road investment are of two kinds:

• Transport cost savings for the existing goods transport
• Industrial Re-Organization Benefits

Mohring & Williamson (1969) concluded that the latter benefits constitute only a small fraction- about 10%- of the total benefits of road investments. By the present model, a similar result is obtained for a cost-minimizing multi-plant monopolist. In a competitive market, which is not studied in Mohring & Williamson (1969), the present model predicts the Industrial Re-Organization Benefits to be some five times larger - and the "welfare triangle" represents only a fraction thereof.