RADIO TRANSMISSION AND SOLAR PHENOMENA 159 



been correct 71 per cent of the time. These figures have been de- 

 termined solely on the basis of "disturbed" or "undisturbed" and 

 modification of the forecasts by the words "probably" and "possibly" 

 have not been taken into account. 



This method of making predictions, even in its present state, is of 

 definite use commercially. Special forecasts of the same nature have 

 proved useful in planning certain experimental studies. 



Well defined sequences of activity of an approximately 27-day period 

 are also apparent in data of solar phenomena, particularly those 

 relating to sunspots, bright and dark hydrogen flocculi,' and promi- 

 nences. An attempt to link such solar sequences with the terrestrial 

 ones noted above in a cause and effect relationship was not, however, 

 very successful. For several well marked radio (and magnetic) 

 sequences it was found that no single type of solar activity could be 

 identified in such a manner as to exhibit a clear cut relationship. 

 For some of the sequences there was some form of solar activity near 

 the center of the sun at the time of each radio disturbance but such 

 activity varied between recurrences in heliographic latitude and 

 longitude and in kind. 



A similar indefinite result was found by starting with the solar se- 

 quences and attempting to match the terrestrial data with them. For 

 instance an area on the sun approximately in heliographic latitude 

 -f 10° and longitude of 315° to 329° exhibited the presence of either 

 hydrogen fiocculi, prominences, or sunspots or combinations of these on 

 each transit across the face of the sun from October 21, 1932 to Febru- 

 ary 9, 1933, a total of five transits. Sunspots appeared in this region 

 on the last four transits and their identity over this period of time was 

 noted at Mt. Wilson Observatory.^ Although the times of central 

 meridian crossing of this area fall within a well defined sequence ^ on 

 the radio chart (between days 6 and 9 on the left at the bottom of the 

 chart) the absence of activity on this solar area for earlier recurrences 

 of the radio sequences tends to vitiate the relationship between radio 

 disturbances and those types of solar activity which were observed. 

 Nevertheless, the reality of the 27-day period in radio is strong indi- 

 cation that solar activity is responsible, even though not convincingly 

 identified in detail. 



Various other criteria were used for segregating the solar data for 

 correlation. For instance, a study of the solar distribution of fiocculi 



' Hydrogen fiocculi are clouds of hydrogen gas observed with a spectrohelioscope 

 set on the hydrogen line Ha. 



^Publ. Astr. Soc. Pac, 45, 263, 53, Feb. 1933. 



* This sequence is considerably strengthened by increasing the number of group 

 ndices into which the data are divided. 



