10 FREER. 



rays in the various portions of the sun's spectrum may vary 

 considerably. 



So much, for the present, for the effect of the more refrangible 

 rays of the sun's spectrum lying in the region of the blue to 

 violet and beyond in the ultra-violet. To them, the greater 

 part of the literature has attributed, in largest measure, the 

 supposed untoward effects of the tropical sun, and to them 

 have been attributed even grave morphologic changes sufficient 

 to bring about permanent differences in races of the human 

 family. So far as the work has gone it seems to develop that, 

 if the so-called "actinic" rays in Manila are particularly objec- 

 tionable, they are the same in Honolulu and for a certain time of 

 the year even in Washington. However, the more we consider 

 the ultra-violet rays of the sun's spectrum, taking cognizance 

 of the fact that nowhere, whether in northern climates or in 

 tropical ones, do they extend beyond 291 p^fi, understanding what 

 a large proportion, if not all, of the direct rays are subjected to 

 molecular scattering, reflection, and dispersion by the upper 

 layers of the atmosphere, and noting the slight differences 

 between the lowlands at Manila and highlands at Baguio, we 

 are forced to the conclusion that, on clear days, when the sun 

 is at the same angle, they are everywhere much alike in intensity. 

 Indeed, it appears as if the greater part of these rays which 

 reach the earth are diffused and not direct. 



These considerations bring us to the much larger remainder 

 of the spectrum which extends upward from the point mentioned 

 into the red and infra-red and which would include the heat 

 rays. That these are a most important factor is, of course, 

 self-evident, and so we, in considering the subject, have not 

 overlooked this fact, but means of direct measurement as in 

 the case of a photocatalytic reaction are lacking. The black-bulb 

 thermometer is variable and unsatisfactory. Better compara- 

 tive data could be obtained if a series of readings of the total 

 solar radiation per square centimeter of surface, normal to the 

 ray of incidence, were available with the Angstrom pyrhelio- 

 meter. 



Such data as are available have been gathered by Dr. Herbert 

 H. Kimball of the Mount Weather Observatory ^^ in a summary 

 which gives the most important figures for the present discussion. 

 Comparisons are made of the annual maximum intensity of solar 

 radiation at various points as follows: 



"^Bull U. S. ML Weath. Obs. (1910), 3, 100. 



