Vector maps are usually contained in a pair of variables, which represent the latitude-longitude pairs for points of interest. For example, the following two variables can be a vector map: Since this is a made-up example, these points could mean anything. Perhaps they are the locations over which geosynchronous satellites are stationed. If so, you might think of them as three distinct points. Perhaps they represent the starting point, the mid-course marker, and the finish point of an airplane race. Think of them as three points connected by two line segments. Or perhaps they represent the vertices of a triangle bounding a region with a propensity for unexplained phenomena, in which case think of them as describing a polygon. The Mapping Toolbox provides functionality for each of these interpretations. For many purposes, the distinction is irrelevant, for others, the choice of a function implies the appropriate interpretation. For example, the command plotm will display the data as a line, while fillm will display it as a filled polygon. When vector variables are used to represent several segments or polygons, successive objects should be delineated by the insertion of NaNs in both variables. For example, if a second segment is to be added to the above map, the two objects can reside in the same pair of variables: Notice that the NaNs must appear in the same locations in both variables. Here we have a segment of three points separated from a segment of four points. The NaNs perform two functions, they provide a simple means of identifying break points in the data, and they act as pen-up commands when plotting vector maps. The NaNs are also used in the Mapping Toolbox to separate non-connected patch faces. Source.