Geomatics, understood as a domain, which scope is defined by international standards of the ISO 19100 group - Geographic Information/Geomatics, covers various aspects of spatial information, but only those aspects which are not directly connected with specific features of a given domain, to which this spatial information refers. For this reason, achievements of geomatics, expressed mostly in the form of abstract conceptual models with regard to geoinformation, cannot be directly used in other domains. In order to use an abstract conceptual model in practice, it is necessary to extend it by elements specific for a given domain - these elements are often called thematic or . more properly . domain-specific. This necessity arises from the fact that abstract models are, in a way, "common denominators", or an abstraction of domain-specific models. For this reason, abstract models contain only such elements, which appear at the same time in many domain-specific application models referring to the same type of information. Domain-specific elements in the abstraction process are omitted. However, there are many cases, when for a specific detailed model in a given domain there is no abstract geomatic model, which could serve as basis for elaboration of a detailed model. Such casus refer to the types of spatial information which appear only in one or in a very limited number of domains and, therefore, elaboration of an abstract model is not justified or not necessary. Geology is a domain for which taking the third dimension (z) into account is necessary and, for this reason, often the types of information used cannot be applied in other domains. It is not exceptional, however, and we may suppose that other domains also use the types of geoinformation not encountered anywhere else and specific only to this domain. Analysis of a few cases with regard to spatial information in geology presented in the paper is aimed at illustrating the problem of domain-specific models. These cases include a geological cross-section in spatial coordinate system (l,z), where the coordinate l is determined along the broken line linking the points of borehole location. Another case as regards geology is a data set concerning a borehole, where one-dimensional spatial coordinate system g (depth from the terrain surface in the point where drilling was started) is determined on a curve, fragments of which may be distinctly not parallel to the perpendicular. Another example refers to description of a process of creating a geological profile in a specific point. In this case, it is necessary to apply a two-dimensional spatio-temporal coordinate system (z,t)] and to use geometric and topological elements adequate for that coordinate system. Particularly difficult problems in this area and specific for geology appear in the cases of registration of spatial models of geological structures. This requires application in these models of three-dimensional coverages (as defined by ISO 19100 standards) and many different types of coverages having that dimensionality may be used. In a structural geological model, three-dimensional coverages are linked by associations to borehole data sets, geophysical data sets and geological cross-section data sets. This considerably complicates conceptual models and, consequently, leads to a need of separate research, specific for geology, concerning issues of spatial information related to three-dimensional features in three-dimensional space. This difference of the problems of spatial information in geology allows us to conclude that there is a need to exclude these issues as a separate section of geomatics named "geological geomatics". One of the most important conclusions arising from the analysis performed is the statement that domainspecific conceptual models concerning geospatial information may and should be solved within individual domains in which they are applied. Numerous examples prove that entrusting specialists from other domains with these tasks does not bring positive results.