GENETICS: GOODSPEED AND CLAUSEN 
333 
day, for air-mass ranges 1.3 to 4; 4 to 12; and 1.3 to 20 air-masses, re- 
spectively. All six solar constant values thus found fall between 1.90 
and 1.95 calories. The smallest air masses, as it happens, yield slightly 
the highest values. We conclude that our previous results have not 
been made too small by neglecting to observe during the time when the 
sun is within 15° of the horizon. 
On July 11, 1914, in cooperation with the United States Weather 
Bureau, a recording pyrheKometer attached to sounding balloons was 
sent up to the altitude of about 24 km., where the barometric pressure 
was 3 cm. of mercury, which is only one twenty-fifth of that which 
prevails at sea level. 
Good records of solar radiation were obtained over a period of more 
than an hour, and including the period when the instrument reached 
highest elevation. The mean value of the best three records made at 
highest altitudes, as reduced to mean solar distance, comes out 1.84 
calories per cm.^ per minute. We believe about 2% should be added 
to represent radiation scattered and absorbed in the atmosphere above 
the level reached, making for the probable value of the solar constant, 
from this day's work, 1.88 calories. This value falls decidedly within 
the range of solar constant values we have observed. We state in con- 
nection v\^ith it the following results which are the highest reliable direct 
observations of solar radiation at the various altitudes, as reduced to 
mean solar distance and vertical sun: 
Mount Mount Manned Free, 
Station Washington Wilson Whitney Balloon Balloon 
Altitude 127 m. 1730 m. 4420 m. 7500 m. 24000 m. 
Barometer 75 cm. 62 cm. 45 cm. 30 cm. 3 cm. 
Radiation 1.58 1.64 1.72 1.755 1.84cal. 
Observer Kimball Abbot Abbot A. Peppier (Smithsonian) 
VARIATION OF FLOWER SIZE IN NICOTIANA 
By T. H. Goodspeed and R. E. Clausen 
DEPARTMENT OF BOTANY. UNIVERSITY OF CAUFORNIA 
Presented to the Academy, April 28, 1915 
During the past five years considerable attention has been given to 
the study of the inheritance of flower-size in Nicotiana at the University 
of California Botanical Gardens. In the course of these investigations 
it has been found that flower size varies markedly under different con- 
ditions attending development. This report is concerned with an 
analysis of some of these conditions, and the bearing of such variations 
on the study of flower-size inheritance in Nicotiana. 
