President Ross:-Not that anybody remembers.

Mr. H. H. Gibson:-I thought that was the intention all through the paper.

President Ross:-There is a great area of land now cultivated that could not be cultivated if the lake were raised; and the people who are making a profit off that land would be, I think, entitled to damages. However, that is a question for the lawyers.

Mr. H. H. Gibson:-They are probably using land that belongs to the Government.

President Ross:-No, this land is above the average high water mark. I do not suppose high water mark would mean the highest level.

Mr. H. H. Gibson:-It would not mean flood water.

President Ross:-That is, water piled up by the action of the wind?

Mr. Gibson:-Yes.

President Ross:-No, it would not mean that. Of course high water mark is very difficult to define.

[This Association is not responsible as a body for the opinions expressed in its Papers by Authors.]

SURVEYING BY PHOTOGRAPHY.

BY J. N. WALLACE, O.L.S., D.L.S.

Considering the fact that Canada possesses in the Rocky Mountains some of the most difficult country in the world to survey, it is strange that even amongst surveyors, there should not be a more general knowledge of the method known as photographic surveying, which is so well adapted to rugged districts. This method is of interest not only to surveyors but to photographers as well, for the latter can fairly claim that by its use surveying has been added to the already long list of sciences which are indebted to the art of photography. There can be little doubt that the general want of knowledge of the method is due to the mistaken idea that its practice is involved in mathematics. No doubt some effort is required at first, to realize the general relations of a photograph to the actual country, but the theoretic knowledge necessary for the practice of surveying by photography is not very extensive nor difficult for a surveyor to acquire. It must not be supposed that proficiency in the practical art is so easily gained, for its difficulties are such that natural aptitude and a large experience in the field are necessary to overcome them.

The use of photography for surveying purposes is much older than is popularly supposed. So far back as 1860, the idea was practically developed by Colonel Laussedat, but its use has been very restricted. Surveys to a small extent have been carried out in several European countries, notably in the Italian Alps, where the lower parts of the mountains have been surveyed by the plane table and the higher altitudes by photography. None of these surveys, however, approach the extent of the Canadian surveys. Under the direction of Mr. Deville, the Surveyor-General of Dominion Lands, who originated the use of the method in Canada, photographic surveys already have covered over 25,000 square miles in the foothills of the Rockies and in the mountains themselves in British Columbia and the Yukon. These were commenced in 1887 and have been ex

tended almost every year since then. The foothills surveys include the area between the prairie and the Rockies, from Calgary southwards to near Macleod, covering about 2,000 square miles of country varying in altitude from 3,600 to 8,000 feet. These surveys are plotted with great detail, and are used in connection with irrigation works. About 20,000 square miles of the boundary country between Canada and Alaska have been surveyed in a preliminary manner, and besides extensive surveys along the railway belt in British Columbia, photography has been applied to the survey of part of the country forming the north boundary of British Columbia, and more recently to the survey of the celebrated Crow's Nest coal area.

The photographic method receives its best application in the survey of mountainous districts, for the very ruggedness of such countries, which proves a barrier to ordinary methods of survey, is itself an advantage when we use photography, as it affords stations from which commanding views can be obtained. The chief characteristic of the method is its ability to make a rapid record of the positions of inaccessible points, and hence the more inaccessible and numerous these points the more fully are its advantages developed. There are, however, cases where it can be well used apart from mountains, when we wish to make a survey of anything lying in one horizontal plane or nearly so, which can be overlooked from suitable stations. Such cases may occur in the survey of a rocky coast-line, the shores of a lake, the windings of a river in a deep valley, etc. In such cases, by means of the perspectometer instrument, we can deduce the horizontal plan from the vertical photograph very easily and rapidly.

In considering the general principles of photographic surveying, it may be well to first of all point out the relation existing between the positions of objects in a country, as seen from a camera station, and the positions of the corresponding points in the photograph. We may suppose a surveying camera, so levelled that the plate is exactly in a vertical plane, to be set up on some mountain peak, and a view to be taken of a range across a valley. We know that an image of the distant scene is formed, by the lens, at a distance behind it equal to its focal length, that is, on the sensitized plate. We also know that the points in this image are so situate that a line drawn from any object to the so-called centre of the camera lens and continued onwards will pass through the corresponding point in

the image. Considering a whole series of such lines drawn from the various objects through the centre of the lens, we may regard the image on the sensitized plate as consisting of the points of intersection on the plate of all the lines of sight from the centre of the lens to the objects in the landscape. We may now suppose a transparent screen to be placed at a distance, in front of the centre of the lens, equal to its focal length, and to be exactly parallel to the sensitized plate. The eye placed at the position of the lens and looking through the screen would see the landscape as it were projected on the screen. Such a picture, being formed by the intersection of the various lines of sight on a plane parallel to the plate and equally distant from the centre of the lens, would therefore be exactly similar to the image on the plate. We could, therefore, replace the picture on the screen by superimposing a finished print taken from the negative. By imagining lines drawn from the centre of the lens to the print so placed on the screen, we can realize how it is that a photograph gives a graphic record of the different lines of sight to all the visible objects, as seen from the camera station.

In order to make use of this record we must have some

basis to work on. In the case of every photograph, in addition to the known focal length, which is constant, we must know three things. Firstly, where is the horizon line, that is, where would a horizontal plane through the centre of the lens have intersected the negative while the photograph was actually being taken. Secondly, we must know where the axis of the lens (that is, "the line of sight" of the camera) intersected the negative. This is called the principal point and is always on the horizon line so long as the plate was exposed vertically. A vertical line across the negative through the principal point is called the principal line. These two lines quarter the negative. As the screen picture is similar and similarly placed to the negative, these lines will occupy similar positions on it. Lastly, we must always know the absolute azimuth of the line of sight of the camera for each exposure. While in reality the picture of the landscape is recorded behind the lens, it is much easier to grasp the general principles if we suppose it to have been recorded on the imaginary screen placed parallel to the plate and in front of the lens.

There are very many forms of cameras for surveying purposes, which may be roughly classified as those which are simply