570 



SCIENCE 



[N. S. Vol. XXV. No. 641 



times been employed for guarding against 

 these distortions, are unnecessary. 



Distortions of Photo graphic Films on 



Glass: Sebastian Albbecht. 



In the winter of 1904 an investigation 

 of the distortions of photographic films on 

 glass plates was undertaken by the writer 

 at the suggestion of Director Campbell 

 and Dr. Perrine. A preliminary investi- 

 gation showed that if ordinary good care is 

 taken in the treatment of the plate during 

 development, fixing, hardening, washing 

 and drying, there will be no general dis- 

 tortions, i. e., no distortions extending over 

 a large area of the film. Accordingly, the 

 work was continued by a method which 

 would more readily determine the char- 

 acter of local distortions. Photographs of 

 spectra with numerous lines of different 

 widths are very well suited for this pur- 

 pose. These spectrograms showed that 

 there were local distortions. These dis- 

 tortions were confined in each case to one 

 or a few adjacent spectrum lines, and very 

 rarely covered an area as much as 0.25 mm. 

 square. The displacements of individual 

 points in the distorted area occasionally 

 amounted to 0.02 mm., which on the Cross- 

 ley star photographs is equivalent to about 

 one second of arc, and on the spectrograms 

 taken with the one-prism spectrograph of 

 the Lick Observatory to a radial velocity 

 of about 80 km. per second (for the H_^ 

 region). Such large displacements as 

 these are, however, extremely rare. 



The distortions seemed to be of two dif- 

 ferent kinds : one caused by an actual move- 

 ment of portions of the film, and the other 

 by a displacement of the image apparently 

 due to local differences in the sensitiveness 

 of the film. Some of the characteristic 

 shapes of the distorted lines due to the 

 first cause were: a sine-curve, a question 

 mark, an irregular crinkle, an abrupt bend, 

 a gradual bend which was sometimes 



shared by two and occasionally by three 

 lines, while the lines immediately on either 

 side of the distorted ones were perfectly 

 straight. 



The structure and chemical composition 

 of the film are materially altered in the 

 process of development and fixing, so that 

 the strains produced in the subsequent 

 drying need not be exactly the same every- 

 where in the film as they were before de- 

 velopment. In fact, slight local readjust- 

 ments might be expected, and it is remark- 

 able that the resulting movements in the 

 film should be so few in number, and be 

 confined to such minute areas. 

 A Biefler Clock and a 8 elf -registering 



Right Ascension Micrometer: W. S. 



ElCHELBEEGER. 



In September, 1903, there was installed 

 at the U. S. Naval Observatory one of 

 Riefler's sidereal clocks. No. 70, with a 

 nickel-steel pendulum and mounted in an 

 air-tight glass case. At the same time the 

 following scheme of observing was under- 

 taken with the 9-inch transit circle; one 

 astronomer would observe fundamental 

 and miscellaneous stars for about two 

 hours before sunrise, would observe the 

 day planets and the sun, and would finish 

 his tour of duty with another two hours' 

 star observing just after sunset; a second 

 observer would then commence and do star 

 work for about six hours ; and then a third 

 man would observe a list of stars for about 

 two hours before sunrise. Thus there were 

 frequently three clock corrections deter- 

 mined during the same night by three dif- 

 ferent observers. Each night from Sep- 

 tember, 1903, to August, 1904, on which 

 two or three men observed, has been util- 

 ized for determining the relative personal 

 equations of the various observers. This 

 interval was divided into five parts and an 

 independent determination of the relative 

 personal equation was obtained for each 



