THE STUDY OF TROPICAL SUNLIGHT. 5 



neglected. The most serious difficulty is the temperature co- 

 efficient, which would become apparent in colder climates, thus 

 bringing down the average during the winter months, but where 

 the temperature is known this can be compensated for by cal- 

 culation ; the flasks can be calibrated, the exposure made as far 

 as possible from disturbing influences, and so, comparative meas- 

 urements conducted with a sufficient degree of accuracy to give 

 us a relative knowledge of the total influence of the rays of the 

 sun in the more refrangible portions of the spectrum. If there 

 are great contrasts between various regions, they will be ap- 

 parent despite any errors in the method. 



The basis of investigation having been determined, in addition 

 to arranging for the carrying on of daily observations in Manila, 

 I asked colleagues in various parts of the world to cooperate by 

 a series of measurements with calibrated flasks and standard 

 solutions exposed on a dull black surface, free from buildings, 

 between the hours of 9 and 12. Returns from all the places 

 cooperating are not yet at hand, but, so far results ^'^ can be 

 reported from Kuala Lumpur ^' (latitude 3° 10' north) ; Hono- 

 lulu '' (latitude 21 M8' north) ; Washirjgton '"- (latitude 38° 59' 

 north) ; Tucson, Arizona ^* (latitude 32° 12' north) and Khar- 

 toum, Egypt '-^ (latitude IS'' 36' north) . 



In Manila (Table I) the average per cent of oxalic acid de- 

 composed for one hour during one year was 12.45, with a max- 

 imum of 17.8 for the highest observed day and a minimum of 

 1.15. The average of all days above the general mean was 14.65 

 and below 9.64. Strange to say, the lower average in Manila 

 did not fall during the rainy months of July to October, but oc- 

 curred in November, and the clear months of January, February, 



" The figaires presented in this paper are for some of the places subject 

 to reinvestigation with standardized flasks and solutions, but, in time, as 

 the work progresses, a large range of latitudes will be covered by exact 

 comparative measurements. 



" Through the kindness of the Institute for Medical Research, Dr. Henry 

 M. Fraser, director, Mr. M. BarrowclifF making the titrations; a quartz 

 flask was used. 



" Through the kindness of the Hawaiian Agricultural Experiment Station, 

 Dr. E. V. Wilcox, in charge, Mr. W. T. McGeorge making the titrations, 

 using a 200-cubic-centimeter Erlenmeyer flask which has been sent to this 

 laboratory for standardization since the above was written. 



" Through the kindness of Dr. Raymond F. Bacon, Bureau of Chemistry. 



" Through the kindness of Dr. H. Spoehr, Desert Laboratory. 



" Through the kindness of the Wellcome Research Laboratories, Dr. 

 Andrew Balfour, director, Dr. W. Beam, chemist. 



