June 2, 1904] 



NA TURE 



105 



lasting six montlis. Five tapes were made use of, all 

 five being tested over one kilometre of distance to 

 determine their relative equations. These may be 

 expressed by 1/690,000 maximum and 1/1,200,000 

 minimum of probable error. Altogether more than 

 69 kilometres of base measurement were effected 

 at a cost of 160 dollars per kilo. Commandant 

 Bourgeois maintains that the limits of probable 

 error in linear measurement are in satisfactory 

 relation to the limits of probable error in the 

 angular measurements of the instrument used for tri- 

 angulation. But he does not fully describe the latter. 

 One of the essential features in modern principal trian- 

 gulation is the employment of instruments of half the 

 size and about one quarter the weight of those which 

 were deemed necessarj' twenty-five years ago. Im- 

 provements in graduation and, above all, the introduc- 

 tion of the micrometer eye-piece have so far added to 

 the accuracy of modern theodolites that a 12-inch in- 

 strunvn; in Inrlin rr,. t.-.i^-r-s the place of the jj-inch 



Fig. 2.— Apparatus of M. Hecker for measuring the intensity of gravity in 

 the open sea by c >mparison of the readings of the barometer (i and 2) 

 and hypsometer (3). 



and 36-inch instruments formerly used. Surveyors will 

 probably have their own opinions as to the methods 

 of observation indicated by Commandant Bourgeois. 

 The German method approved by him, and adopted in 

 France,^ appears to contemplate' certain irregularities 

 in the signals for observation which ought not to exist. 

 It will probably be found that the system of observing 

 should be adapted to the atmospheric peculiarities of 

 the district in which the observations are taken. But 

 the German method is well worth the careful attention 

 of English surveyors. 



That part of the article which deals with the de- 

 flection of the plumb line and the intensity of the force 

 of gravity, has a most interesting reference to M. 

 Hecker's apparatus for investigating these problems in 

 ocean spaces by means of a comparison between baro- 

 metric and hypsometric observations ; the general result 

 of such observations taken in the .-Atlantic between 

 NO. 1805, VOL. 70] 



Bahia and Lisbon being to prove that there is no great 

 variation between the results determined in the deep sea 

 and on the Continent. M. Hecker is still engaged in 

 this branch of geodetic inquiry. 



The reference to the African are now contemplated, 

 and to an equatorial arc recently measured by French 

 scientists in the Republic'of Equador in South America, 

 should be studied together, for the experience obtained 

 in the latter points some useful morals for the consider- 

 ation of those who may undertake the measuretnent of 

 the former. The physical conditions of the country and 

 the variations of an unusually tempestuous season pre- 

 sented but small obstruction to the progress of the work 

 compared to the hostility of the indigenous Indians. 

 Stations were destroyed and markstones uprooted with 

 such persistent animosity in Equador that a great part 

 of the observations had to be repeated. If principal, 

 or geodetic, triangulation is to serve the purpose of 

 scientific investigation only, the destruction of the 

 observing stations would not be of so much conse- 

 quence, when once the chain of triangles composing 

 the arc was finally complete. But it is obvious that if 

 any useful ulterior purpose of map-making is to be 

 served by the expensive process of laying down a back- 

 bone of well-fixed points, it is all important that every 

 station and every markstone should be preserved with 

 the utmost care. In spite of most elaborate precautions 

 these most necessary indications are sometimes lost in 

 India, and fresh observations have to be made in order 

 to redetermine their position. Isolation of the instru- 

 ment during the process of observing is almost always 

 imperative, although it occasionally happens that a con- 

 siderable area of hard rock exists of sufficient stability 

 to serve as the basis of the observing station without 

 mvolving any artificial isolation. But the building of 

 isolating pillars and the erection of cairns over them for 

 protection almost inevitably attracts the attention of the 

 tribespeople in the neighbourhood, and the result is 

 subsequent destruction. 



The only way to safeguard with any prospect of suc- 

 cess against the utter waste of time and money which 

 IS involved by the destruction of signals and mark- 

 stones, after the triangulation has been effected with 

 scientific precision and rigorous methods of observation, 

 IS to fix, pari passu with the principal triangulation, a 

 large number of secondary points scattered over the 

 face of the country, consisting of natural features which 

 It is impossible to remove, or for Indians to identify. It 

 cannot but happen that principal triangulation carried 

 through an arc of 65° of amplitude in such a 

 country as .Africa will involve a great deal of native 

 hostility, and its preservation finally will be almost an 

 impossibility. It will be most necessary, therefore, to 

 take all classes of observations that have eventuallv to 

 be taken from any one station at one and the same time 

 of occupation. It may indeed be an open question 

 whether one or two short principal series from the coast 

 westward, following, say, the Zambesi and the Uganda 

 Railway to the meridian of 30° E., would not "suffi- 

 ciently answer the utilitarian purposes of a basis for 

 African surveys were they connected by secondary or 

 even tertiary triangulation at their extremities, and' the 

 connection pushed northward to meet a third principal 

 series on the Nile. This, however, is but a side issue 

 prompted by the perusal of the admirable article in the 

 Revue Generale des Sciences. 



One especially interesting result of the observations 

 for level deflection taken in connection with the Equa- 

 dor arc, is an indication that the compensation of ex- 

 terior mass by interior deficiency, or want of density, 

 indicated by sucTi observations at certain Himalayan 

 stations, does not exist in the equatorial region of the 

 Andes. T. H. H. 



