October 10, 1913] 



SCIENCE 



515 



Coast and Geodetic Survey. Washington, 



Government Printing Office. 1913. 



It is the purpose of the reviewer to discuss 

 Parts I. and II. only — the parts relating to 

 the determination of time and longitude. 



The reason for the appearance of the volume 

 is twofold: first, the fourth edition, by Pro- 

 fessor John F. Hayford, at the time inspector 

 of geodetic work, has become exhausted; and 

 secondly, so much that is new has developed in 

 the interim, and so much of the old has be- 

 come changed or entirely discarded, that it has 

 been thought advisable, even though much of 

 the old material may be found scattered 

 through other publications, to issue still an- 

 other volume, one which shall be in itself 

 complete and thoroughly accordant with pres- 

 ent practise. So great is the demand for this 

 valuable manual that the new edition has 

 already nearly given out, and it has conse- 

 quently been found necessary to order the 

 printing of an additional thousand copies. 



The self -registering transit micrometer, in- 

 troduced by Repsold a quarter century ago, 

 and the principle soon after adopted in Europe 

 of reversing the transit instrument during the 

 observation of each star, have almost revolu- 

 tionized the methods of longitude determina- 

 tions. Their advantage is that they afford 

 additional strides forward in the direction of 

 eliminating constant and systematic errors by 

 skilful observational manipulation rather than 

 by applying corrections in the course of the 

 computation. Reversal during the transit of 

 each star eliminates collimation, inequality of 

 pivots, irregularity and other errors of the 

 transit micrometer screw (or, if a fixed reticle 

 be employed, the thread intervals), and, in the 

 ease of the broken-back telescoi>e, bisection 

 error and flexure. With regard to the mi- 

 crometer, though claim has been made that its 

 use leads to a higher degree of precision, its 

 chief value lies in that it almost annihilates 

 the observer's personal equation. As the in- 

 strumental and personal equations are thus so 

 greatly reduced, further approach is rendered 

 possible towards the ideal arrangement of re- 

 ducing the observational errors exclusively to 

 the accidental type. These two innovations 



have accordingly been attended with so great 

 success that, employed originally in the field, 

 they have found their way even into the fixed 

 observatory. 



There the right-ascension micrometer has 

 come to stay. Whether the ponderable tele- 

 scopes of the fijsed observatory can be adapted 

 to quick reversal, however, remains yet to be 

 seen. At Kiel and at Bergedorf they are 

 employing transit circles designed and built 

 with this purpose in view, but the onlooker 

 during the operation of reversal instinctively 

 fears for the safety of the instrument. Ex- 

 periments are still under way. Speedy re- 

 versal with the portable transit, on the other 

 hand, was long ago effected by both the Ger- 

 mans and the French, the latter developing 

 the straight telescope with diagonal eye-piece, 

 the former the broken back. 



Ever alert as the Coast and Geodetic Survey 

 authorities are for any device bearing the 

 impress of improvement, they have stamped 

 their mark of approval upon both these inno- 

 vations. Though they have not purchased or 

 made any astronomic instruments for time 

 observations since the appearance of the fourth 

 edition of their manual, they have recently 

 ordered two telescopes of the broken type, 

 reversible on each star, such as have " been 

 used with marked success by other countries," 

 and illustrated in Plate 2. The right-ascen- 

 sion micrometer they welcomed a decade back. 

 Skilfully designed by the chief of the instru- 

 ment division, Mr. E. G. Fischer, and tested 

 in a thorough experimental and theoretical 

 investigation by Professor Hayford,^ the mi- 

 crometer has since proved of such worth that 

 the effect of its introduction may be traced 

 throughout the new edition. 



When the chronographio method of star 

 registration was introduced in the middle of 

 the last century — and it will be recalled what 

 a prominent part the Coast Survey played in 

 the introduction — astronomers fondly hoped 

 to eliminate by its means that most trouble- 

 some of " constant " errors, personal equation. 

 That it greatly reduced the magnitude of this 



^ Appendix No. 8, Report of the Superintendent 

 for 1904, "A Test of a Transit Micrometer." 



