Abstract T he I nstitute of G eological S ciences is increasingly being asked to provide geological information along the lines of proposed motorways and major road improvements. It is rare, however, for such requests to be received before the line survey is completed, as was suggested in the then Ministry of Transport's Technical Memorandum T1/67. Consequently, advice tends to be limited to comment on likely difficulties along a route already established. This procedure works well in areas where geological sequences and structures are simple and consequently where geological factors are greatly outweighed by social factors. As motorways push into more difficult geological terrain, however, geological factors become increasingly more significant and an early approach to the Institute correspondingly more desirable. It is general for the Institute to re-survey proposed road lines on request and bring older mapping up to modern standards. Moreover, collection of geological data in any regional study is necessarily selective and by up-dating the mapping with a specific project in mind, observations pertinent to that project can be made that might otherwise be considered too time-consuming to be justifiable, and the less obvious implications of the mapping can be clarified. This sort of work can clearly be best done by the Institute with its pool of experienced field staff and its comprehensive back-up facilities. In contrast, many of the problems that arise during construction are essentially local ones and could be dealt with adequately by a site geologist using the Institute staff only where necessary. If geology is to be properly used at this stage some such arrangement must be made, for the field staff of the Institute is not large enough to deal with such matters on a general basis. However close co-operation is between geologists and engineers, there is still a significant difference in approach between the two disciplines. The engineer requires quantification of rock properties so that he can calculate the effects of the interplay between his constructions and his site; the geologist doubts the validity of many such quantifications, at least in non-isotropic material intersected by a variety of differently inclined weakness planes, and would prefer to use an empirical approach, anticipating likely results by studying the history of comparable constructions in similar geological environments. To do this properly much more geological research is needed into the stability of natural slopes and established cuttings and embankments, whether or not these have histories of failure, and into devising with the engineer effective experiments on in situ rocks. In any such work the field staff of the Institute should be fully involved, so that the results obtained can help in making its primary route assessments more meaningful to the engineer.