TRANSACTIONS OF SECTION RF, 451 
settings of the circle are quite sufficient to eliminate in their mean all appre- 
ciable periodic errors. 
In the paper already mentioned as having been read to the Australasian 
Association in 1898 the writer referred to an apparent connection between the 
heights of the lines above the intervening surface and the values for the coefti- 
cient of refraction derived from the reciprocally observed zenith distances. 
The extension of the survey since 1898 has given further data which generally 
bear out the conclusions then derived. These data and the results are dealt 
with in the present paper. In the absence of a topographical survey from 
which to ascertain the height of each line above the intervening surface the 
observations have been grouped in the order of length of line, the assumption 
being that generally the longer lines are the higher. Another grouping has been 
made in which account has been taken of the height above sea-level on the 
assumption that the higher regions are the more hilly and that there the height 
of the ray above surface is the greater. Both groupings would seem to indicate 
that where information as to the highest above surface is lacking varying 
coefficients of refraction may be assumed according to the lengths of the lines 
observed. Diagrams are given showing the variations. 
The survey has been computed from the Sydney Observatory (Lat. 
33° 51/ 41/-1 §.; Long. 151° 12’ 23/-1 EK.) as origin. A map accompanying 
the Paper shows the extension of the survey therefrom southerly about 240 miles 
to the Victorian border, where it connects with the triangulation of Victoria, 
westerly about 360 miles to the limits of the almost flat country of the interior, 
and north-westerly towards the Queensland border a distance of 240 miles. 
The map shows also the differences between the geodetic latitudes, longitudes, 
and azimuths as derived from the assumed earth dimensions (a modification of 
the Clarke 1880 spheroid) and the corresponding latitudes &c. obtained by 
astronomical observation. These show the effects of local deflection of the 
vertical caused by irregular distribution of surface-masses. Broadly, New South 
Wales may for our present purposes be said to be divisible into three zones, the 
littoral of from 20 to 60 miles in width rising from sea-level to 2,000 feet in 
height ; at the rear of that plateau 100 miles in width varying from 2,000 to 3,000 
feet in height, with mountain masses up to as much as (in the southern extremity) 
7,000 feet, and westward of that a more or less gentle western slope to interior 
plain country. As the result of the attraction of the central elevated mass and 
of the defect of gravity of the adjacent ocean it would appear that along the 
coast there is a general eastward deflection of the zenith of about 10/ with a 
corresponding westward deflection on the western slopes ranging up to as much 
as 17’, but gradually diminishing as the flat country is reached. In 1898 a 
general reduction of the data then available was made for the purpose of obtain- 
ing an idea how far there was conformity between the actual surface and the 
assumed spheroid, but the survey was then too much limited to the eastern 
slopes to afford satisfactory evidence. With the subsequent extensions of the 
survey, however, a much more useful discussion of the subject is within reach, 
and it was hoped that by now bases of’ verification on the outskirts of the work 
would have been measured, when it would have been possible to reduce the 
whole work with the object of enabling such a discussion to be made. 
2. The Sand-Drift Problem on the Eastern Coast of Australia. By 
G. H. Hauniaan, F.G.S., Inspecting Engineer and Hydrographer, 
N.S.W. 
The scientific, as well as the commercial, importance of a full knowledge 
of the direction, volume, and velocity of the movement of sand on the eastern 
coast of this continent is sufficient warrant for the labour and cost expended 
upon it by the author during the last thirty years. 
There is undoubted evidence of the sinking of the eastern coast of Australia 
during recent geological time to the extent of 200 to 300 feet, and the natural 
result was to leave a very uneven shore-line, with many outlying islands, deep 
bays, and rocky capes. Had there been no ocean current, running parallel 
to the general trend of the shore, the sand and shingle, resulting from the 
GG2 
