January 7, 1909J 



NA JURE 



285 



A glance at the curves will show that the sanv' 

 declination is occupied by different stars at different 

 dates; hence it may happen that the declination found 

 tits more than one star wilhin probable date limits, and 

 so we have to decide which is the more likely star to 

 have been obser\-ed. ll might at first sight seem that 

 it would be difficult to settle which star is really in 

 question. But in practice the difiiculty does not often 

 arise. We now know that the stars used vt-ere those 

 in high northern or southern declinations for noting 

 the time at night in the way the Egyptian temples 

 have familiarised us with, and stars nearer the equator 

 to serve as " morning stars," warners of sunrise. 



The stars with about the dates already revealed by 

 the work of the last few years may certainly be con- 

 sidered in the first instance. 



It is really not a remarkable fact that so few stars 

 are in question, for the use made of them was very 

 definite. Capella, .Arcturus, o Capricorni, Pleiades, 

 and .\ntares almost exhaust the list. 



The use of the precessional globe saves many intri- 

 cate and laborious calculations when only an ap- 

 proximation is required. Thus warning stars at any 

 quarter of the Mav or solstitial year at any given 

 date may be found by rectifying the globe for the 

 latitude of the place of observation, marking the 

 equator at that date by a circle of water-colour paint 

 by holding a camel's-hair pencil at the east point of 

 the wooden horizon, and rotating the globe. The in- 

 tersection of the equator and the ecliptic gives us the 

 equinoxes at that date, their greatest separation the 

 solstices. With these data we can mark the required 

 position of the sun on the ecliptic. 



This done, if we rotate the globe so as to bring the 

 sun's place io° below the upper surface of the wooden 

 horizon, the star the rising of which can be used as ;i 

 Warner will be seen on the horizon. 



Nor does the use of the globe end here. With, a 

 given azimuth, which are all marked on the wooden 

 horizon, the globe may be adjusted to different dates 

 and then rotated until at a certain date a star rises at 

 that .izimuth. 



Norman Lockver. 



GEODETIC SURVEYS. 

 T^HE latest volume (vol. xviii.) of the Great Trigo- 

 -'• nometrjcal Survey of India contains the records 

 of astronomical observations for latitude extending 

 over the last twenty years. It is, in effect, the con- 

 tinuation of vol. xi., and brings this particular de- 

 partment of Indian Survey statistics up to date. It 

 consists chiefly of tabulated records ; 543 pages alone 

 in part ii. being absorbed by tables of astronomical 

 latitudes. There is therefore nothing to offer in the 

 way of remark or criticism on the great bulk of detail 

 contained in this volume except congratulation on the 

 comjjletion of a work involving so much labour in 

 compilation. It is, perhaps, the most interesting ot 

 the whole series of Great Trigonometrical Survey 

 records and the interest of it to the general reader lies 

 in the preface, where Colonel Burrard, in plain and 

 simple language, gives the reasons for the faith that 

 is in him as regards the present position of geodetic 

 v\ork in India. 



To those who have pinned their faith to the rigid 

 accuracy of geodetic triangulation as the basis of 

 fixed points for the further extension of minor systems 

 of triangulation and of topographical survey, it may 

 at first sight appear somewhat disturbing to be 

 assured that there is no finalitv in sight for the value 

 of any fixed point in India, either in latitude, longi- 

 tude or altitude. Geodetic science can onlv develop 

 on a system of trial and error. Only by the most ^ 



NO. 2045, VOL. 79] 



rigidly exact systems of measurement possible to 

 human agency can the shape of the earth's figure be 

 precisely determined, and only, when the precise shape 

 of that figure has been determined, can geodetic cal- 

 culations be satisfactorily computed. Hitherto these 

 calculations in India have been based on an assumed 

 earth-figure known as Everest's spheroid, and al- 

 though this assumption is not absolutely justified by 

 continuous observation, Col. Burrard rightlv main- 

 tains that it would be a mistake to break the con- 

 tinuity (and thereby destroy much of the value) of thi^ 

 (ireat Trigonometrical Survey series by the introduc- 

 tion of tables based on new, and possiblv only 

 half-digested, data. Similarly he pleads for absolute 

 accuracy in the determination of latitudes, for it is 

 only when the riddle of the earth's shape shall be 

 solved by the men of science of the future, and the 

 pathway to positive deductions therefrom straightened 

 out, that the full value of this most remarkable body 

 of results (obtained by new and more perfect instru- 

 ments from observations of stars of which the position 

 is now more certainly known than heretofore) can be 

 effectively utilised 



The deflection of the plumb-line forms one of the 

 principal subjects of scientific investigation of which 

 the record is to be found in this book. This deflec- 

 tion is determined by the dift'erence in latitude ob- 

 tained for any fixed point between the results of 

 geodetic triangulation and of astronomical observa- 

 tion. For reasons already suggested in connection 

 with the assumption of the earth's figure, as well as 

 the fact that the origin of geodetic latitudes in India 

 (at the Kalianpur base) is itself an assumption, there 

 still remains an element of uncertainty in these deter- 

 minations. They are exceedingly interesting. " In 

 the Himalayas " (which is, perhaps, a slightlv vague 

 definition) the deflection amounts to —35.29s. ; at the 

 foot of tht Himalayas it is — lo.gos. ; in central India 

 it amounts to -I-0-94S. But it must be remembered 

 that in dealing with this matter of rigid accuracy we 

 have still to reckon with minutely small errors, quan- 

 tities that are immaterial for the practical purpose of 

 supplying a basis for map-making. For instance, the 

 most improved methods of observing with the best of 

 new instruments only displaces the assumed value of 

 the Kalianpur latitude by 0-3". In the matter of 

 longitude there is, however, a recognised error of 

 2' 27", which is an error too large to be neglected. 

 This has to be eliminated from Indian mapping; 

 although, again. Col. Burrard deprecates any inter- 

 ference with the continuity of Great Trigonometrical 

 Survey records in the series ended by this eighteenth 

 volume. To this extent Indian topography and Indian 

 geodesy must remain discrepant for a space of time. 



There is, however, one element of disruption in 

 Indian Geodetic Survey work with which no man of 

 science can deal. This is caused by earthquakes, and 

 the resulting displacement of mark-stones is not easily 

 determined. There may be little relative displacement 

 over a large area, whilst the absolute displacement of 

 the whole area mav be considerable. It is impossible 

 to re-triangulate the vast spaces which would be 

 necessary to determine this, nor does it appear to be 

 at all easy to discover what may be the efTect of such 

 disturbances in altitude. The most careful levelling 

 (three times repeated) over the eighteen miles separat- 

 ing Dehra from Mussoorie only revealed a probable 

 diminution of 55 inches in the Himalayan altitudes at 

 Mussoorie after the latest, and most violent, earth- 

 quake. Meanwhile geodetic science fulfils its mission 

 admirably in the great practical work of establishing 

 the basis for topographical survevs. These never can 

 be affected bv those small geodetic adjustments which 

 are all-important to the scientific theorist, although it 



