212 ANNUAL EEPORT SMITHSONIAlSr INSTITUTION, 193 5 



illustrated in plate 1, figure 2, seem to present such a possibility. 

 Showers may be recorded to the exclusion of all other rays if use is 

 made of their divergence in angle from the point of origin. Three 

 counters arranged in a triangle and surmounted by a block of lead 

 record coincidences only when a shower is produced in the lead. 

 Several investigations have shown that the shower particles are gen- 

 erated in the lead by impact of a specific form of radiation, probably 

 photons. If these photons are generated by primary electrons, as the 

 theory indicates, and not by protons, the showers can be used as a 

 measure of the electrons to the exclusion of the protons. During the 

 directional distribution survey, showers were also studied in relation 

 to changes of latitude, elevation, and direction, and the results are 

 suggestive of electron primaries. It was found that the showers 



increase more rapidly with elevation 

 than the total radiation, though in this 

 respect they cannot be definitely distin- 

 guished from the unbalanced positive 

 component which gives rise to the asym- 

 metry. In fact the results at first sug- 

 gested that the showers were closely 

 associated with the positive component. 

 Measurements in Mexico of the depend- 

 ence of shower intensities on azimuth 

 have now shown that this is not the case. 

 -Arrangement of three Using an arrangement of counters illus- 

 and a lead block for trated in figure 7, showcr intensities from 



cosmic-ray showers <= ' 



due to primary rays from par- the east and west wcre Compared and the 

 ticuiar directions. ^.^^^^j^g showed almost no asymmetry. 



Latitude intensity variations of this component on the other hand have 

 proved that the showers are caused by electrically charged primaries. 

 There is only -one conclusion. The primaries which produce the 

 showers consist of equal numbers of positive and negative rays. Until 

 there is more evidence for the existence of the negative proton, and 

 in view of the ability of these rays to produce the shower-generating 

 photons, we must regard this component of the primary radiation as 

 an electron component. The equality in number of the positives and 

 negatives is also agreeable to this view for the cloud-chamber experi- 

 ments of Anderson have shown that electrons often appear as paired 

 positives and negatives. In spite of the evidence that much of the 

 primary charged component is positive, there is still place to fit in 

 a small balanced component of positives and negatives, particularly 

 at high elevations. It may also be true that as far as the effects 

 recorded by aligned counters is concerned the electron component is 

 quite insignificant. 



