Ordnance Survey – Great Britain's national mapping agency

The Cassini Projection

The Cassini Projection (or Cassini -Soldner) was the projection most commonly used by Ordnance Survey until 1945.

It may be defined as a simple non-perspective cylindrical projection.

Its most important property is that scale along its central meridian, and everywhere at right angles to it, is true. Elsewhere on the other hand, the projection distorts the scale in a roughly north to south direction, by an amount which varies as the square of the distance from the central meridian; the greater the longitudinal extent, therefore, the less suitable is Cassini's projection likely to be.

As far as the Ordnance Survey is concerned the complexity arises not so much from the general adoption of this simple projection as from the mode of its application. The essence of Ordnance Survey practice throughout the nineteenth century was that there was not one national Cassini projection covering the whole country, but a series of independent Cassini projections relating either to a county or a group of counties. Such an approach to map projections in a national survey was widely accepted. A commonly held view in the nineteenth century was that maps of small districts should be complete in themselves: large scale surveys, not only in Great Britain, but also in countries such as France and Belgium, were plotted on a series of separate meridians. It was recognised that this could give accurate results for the purpose for which the maps were designed.

The Cassini projection, introduced in 1745, was first used by Ordnance Survey in the Old Series one-inch maps of England and Wales (1805-73). Although it has been claimed that the one-inch was projected on a single Cassini projection, 'so as to form a continuous series covering the whole of England and Wales', * this is unlikely. Winterbotham accepted the nineteenth century view that there were six central meridians for the old one-inch, stretching from Greenwich to Devon, but a recent study of the Old Series suggests that only two central meridians, probably those of Butterton Hill and Greenwich, were actually used in the construction of the maps of southern England and, in northern England, they were projected on the meridian of Delamere.

In the survey of Ireland (from 1825 onwards), a system of local Cassini projections in which the six-inch maps of any one county were plotted independently of those of other counties was also adopted. A point near the centre of the particular county was selected="selected" and along the meridian passing through it lay the origin of the projection relating to that county.

It was this technique which, after 1840, was transferred into the large scale survey of Great Britain. To a certain extent it was an unavoidable expedient because, when large scale surveys were put in hand, the selection and adjustment of the primary and secondary triangulation for the whole country still had to be completed and was not fully available to the detailed survey.

As a result, a succession of local central meridians were brought into use before the associated tertiary triangulation could be adjusted to the primary triangulation of Great Britain. They proliferated especially before the 'Battle of the Scales' was resolved, and the large scale maps of most of the counties in southern England, and also of many towns, were published on their own Cassini projections.

During World War 1 the Cassini Projection proved to be unsuitable for artillery purposes because of the distorted scale. The possibility of using a conformal projection was therefore investigated and beginning in 1931 with the fifth edition, the 1:63 360 scale map of Great Britain was redrawn on the Transverse Mercator projection which by that date had already been used in a number of colonial surveys.

Source: J B Harley - Ordnance Survey Maps A descriptive manual - 1975
* J. A. Steers, 225; chapter XII dealt with 'Ordnance Survey map projections and grids'.


Search for more GeoFacts

  Advanced  | Help


All reasonable steps have been taken to ensure that the information contained in these Geofacts is correct; Ordnance Survey does not warrant their accuracy. Please independently verify their accuracy before relying on them.

Top of page