December 23, 1921] 



SCIENCE 



637 



Regardless of what the latter had been, cultures 

 which contained healthy and extensive growth 

 tended to be neutral, those which failed to grow 

 had usually become slightly acid, and those 

 that had exhibited extensive growth and then 

 degenerated were most frequently slightly alka- 

 line. These results, however, apply only to 

 solutions containing not more than 0.5 per 

 cent, dextrose, for when 1 per cent, or more 

 dextrose was added to the medium the cultures 

 were often found to be acid when death took 

 place. 



In these observations the optimum hydrogen- 

 ion concentration for tissue cultures in Locke- 

 Lewis solution was pH 7. The final concentra- 

 tion depended upon the amount of dextrose in 

 the medium. Cultures in media containing no 

 dextrose usually had a hydrogen-ion concen- 

 tration ranging from 7 to 7.6; those in media 

 having 0.25 to 0.5 per cent, dextrose ranged 

 between pH 6 and pH 7.8, mostly pH 7.2 and 

 pH 7.4; while those in media to which 3 per 

 cent, and 5 per cent, dextrose had been added 

 were often pH 6 and pH 5.6 respectively. 

 M. E. Lewis, 

 Lloyd D. Felton 

 The Johns Hopkins Medical School 



AN ELECTRICAL EFFECT OF THE AURORA 



During the past year I have been making 

 observations on the diurnal variation in 

 electric potential difference between the earth, 

 as represented by the water system of Palo 

 Alto, and an uncharged, insulated conductor 

 kept inside an earthed metal cage. The records 

 of this variation have been registered con- 

 tinuously by a photographic method since 

 July 20, 1920. For two weeks, or more, 

 preceding the great aurora of May 14 these 

 records were different from any which had 

 preceded them, and two days before the be- 

 ginning of the aurora there was a sudden 

 change in the potential difference being 

 measured which seemed to indicate an in- 

 crease in the negative charge of the earth. 



After the aurora the record of the diurnal 

 variation was of a very different character 

 from anything which had been obtained be- 



fore. In Fig. 1, the continuous line repre- 

 sents the mean variation of the recorded 

 potential-difference in millivolts for ten days 

 preceding the aurora, and the broken line 

 gives the same data for the ten days follow- 

 ing the aurora. The mean daily range of the 

 recorded potential difference on my record 

 was 99.5 millimeters for the ten days preced- 

 ing the aurora and 35.5 millimeters for the 

 same period following the aurora. 



Fig. 1. Diurnal variation in potential differ- 

 ence between the earth and an uncharged, insu- 

 lated conductor for ten days preceding and ten 

 days following the aurora of May 14, 1921. 



The mean diurnal variation in millivolts 

 for the ten months, August, 1920, to May, 

 1921, is shown by the curve in Fig. 2. 



Fig. 



A simultaneous record of the change in 

 the north component of the earth's magnetic 

 field was made on the same sheet with the 

 electrical record. For three days at the time 

 of the aurora the magnetic record was too 

 much disturbed to admit of measurement. 

 The mean range of magnetic variation for 



