INSTRUMENTS, OBSERVATIONAL PROCEDURE, AND CONSTANTS 



15 



Kolhorster, a factor for the former could be found. The 

 number of such series that would be required would de- 

 pend on how consistent the Kolhorster results were for 

 the indoor and outdoor positions. A minimum of four 

 series was planned, each of about four hours duration, 

 for the early part of cruise VH, two of these to be with 

 all sails furled, if possible, and two with all sails set. 

 Definite results from these might obviate further tests. 



During the regular daily observations and during 

 diurnal -variation runs, the Kolhorster instrument waste 

 be placed at a convenient position in the atmospheric- 

 electric house, preferably between the ion counter and 

 PRl. It was desired particularly that simultaneous ob- 

 servations be obtained with PRl and Kolhorster 5503 

 during diurnal-variation observations. 



In accordance with the suggestions of the foregoing 

 paragraphs, simultaneous measurements were made with 

 PRl and 5503 from July 9 to 19, 1928, between Hamburg 

 and Reykjavik, and again after leaving Reykjavik from 

 July 28 to August 18. From these results an attempt was 

 made to determine the residual ionization of PRl, but 

 the scatter of the data for apparatus 5503 was so great 

 as to prevent getting a satisfactory value. The data for 

 the period August 10 to 18 were not used in this inves- 

 tigation, as it was noted that after August 9 a radical 

 change in the measurements with apparatus 5503 oc- 

 curred, much higher values of ionization being obtained 

 with it after that date without correspondingly higher 

 values for PRl. In the progress reports of the next sec- 

 tion (pp. 31-45), difficulties with apparatus 5503 are dis- 

 cussed; continued difficulties culminated in abandonment 

 of the program contemplated for this instrument and the 

 return of the apparatus to Washington for repair when 

 the ship arrived at Canal Zone in October 1928. 



The repaired apparatus was returned to the ship at 

 Tahiti in March 1929. The repairs included the replace- 

 ment of the broken "charging arm" and replacement of 

 the fibers and fiber mounting of the electrometer ele- 

 ment. These changes altered the capacitance to 0.348 

 cm in comparison with 0.374 cm as given by Kolhorster. 

 Both these figures subsequently were corrected when in- 

 formation was received that a correction factor of 1.243 

 should be applied, the new values being 0.433 and 0.465 

 cm, respectively. 



After March 20, 1929, simultaneous measurements 

 with PRl and 5503 were again made, but the scatter of 

 data obtained with 5503 again prevented determination of 

 the residual ionization of PRl. In general, the operation 

 of apparatus 5503 was unsatisfactory throughout cruise 

 Vn, and the data are being omitted from this publication. 

 Unfortunately, this apparatus, as also all other instru- 

 ments and pieces of apparatus, was lost with the ship on 

 November 29, 1929, and therefore no attempt could be 

 made later, in the laboratory, to find the causes for the 

 unsatisfactory performance. 



Penetrating radiation apparatus 1 thus was left 

 without an established value for the residual ionization. 

 Measurements were made regularly, however, through- 

 out cruise VII with this apparatus, the average value 

 for 368 daily measurements being 2.8 pairs of ions per 

 cubic centimeter per second, not corrected for residual. 

 Each daily measurement required between one and two 

 hours and the period of observation was arranged to 

 coincide with the period required for conductivity and 

 ion content measurements. 



Attempts were made on thirty-two occasions to 

 make diurnal-variation observations of penetrating ra- 



diation with PRl; twenty-six series of twenty-four 

 hours each were completed and six left Incomplete. So 

 much scatter was found among the hourly measurements 

 composing these series, that the publication of this 

 material as representative of the diurnal variation of 

 penetrating radiation (or cosmic rays) is scarcely war- 

 ranted. Furthermore, so much has been done since 

 1929 by other investigators toward establishing the 

 character and magnitude of diurnal variation of cosmic 

 rays that little need now exists for the data of cruise 

 VII. Accordingly, this material will not be presented. 



Mauchly (11), dlscussingthe daily measurements of 

 penetrating radiation made with PRl on cruises IV and 

 VI, gave the average values, not corrected for residual, 

 as 3.2 pairs for cruise IV and 3.8 pairs for cruise VI. 

 These means were obtained for observations on 296 and 

 316 days, respectively. On the matter of residual ioni- 

 zation he said " this probably does not much exceed 



two pairs per cubic centimeter per second, since it is 

 not uncommon to observe a total ionization of the order 

 of two and one -half pairs per cubic centimeter per sec- 

 ond, with several extreme cases going even below two 

 pairs." In one of the progress reports of the next 

 section (p. 33), the suggestion is made that lack of a 

 roof aperture over the penetrating radiation apparatus, 

 such as had been used on previous cruises, might be re- 

 sponsible fo." the lower value of ionization (2.8 pairs) 

 observed on cruise VII. This seems unlikely, however, 

 because at sea the penetrating radiation is largely, if not 

 entirely, cosmic rays and the presence or absence of the 

 roof above the apparatus should make little difference. 

 A more likely explanation might be that the lower value 

 of cruise VII was the more accurate value, owing to 

 greater reliability of the apparatus as a result of im- 

 provements incorporated just prior to the start of the 

 cruise. The balancing condenser, in particular, was im- 

 proved considerably. 



Mauchly' s value of two pairs per cubic centimeter 

 per second for the residual ionization of PRl seems 

 somewhat too high for the cruise VII data. A more ap- 

 propriate value would seem to be about 1.5 pairs, thus 

 leaving the balance of 1.3 pairs as the average ionization 

 by cosmic rays. The scatter of the observed valuesfrom 

 the average of 2.8 pairs per cc per second may be indi- 

 cated by stating that 60 per cent of the values fall between 

 2.6 and 3.0, 90 per cent between 2.4 and 3.2, and 95 per 

 cent between 2.2 and 3.4 pairs per cc per second. The 

 individual values will be tabulated in a later section in 

 which the daily observations of the several atmospheric- 

 electric elements will be presented. 



Radioactive Content of the Atmosphere. - -Apparatus 

 for the measurement of the radioactive content of the at- 

 mosphere was installed on the Carnegie before the begin- 

 ning of cruise VII, but successful measurements were 

 not obtained with it until the last two months of the cruise 

 because of persistent insulation failure on one of the three 

 major items of equipment. The apparatus used on cruise 

 vn, as on previous cruises, consisted of a collecting ap- 

 paratus, a high potential generator, and an ionization cham- 

 ber with electrometer attached. The high potential gener- 

 ator was the equipment on which adequate insulation could 

 not be maintained, until the instrument was remodeled. 



The method employed for measuring the radioactive 

 content of the atmosphere consists in drawing air between 

 two concentric cylinders of a collecting apparatus, the 

 central cylinder of which is charged negatively to such a 

 high potential (by the high potential generator), that all 



