wavelengt?! between these two spectra. I would estimate the 



o 



signal to noJ.se at 5500 A to be 20/1. The irradiance varies 

 by about a factor of 4 to 5 betv7eGn these spectra. Most 

 of this variation is probably due to cloud cover changes. I 

 would conclude that there is a higli degree of consistency in 

 spectral structure in the resultant data. Cloud cover variaijon 

 does not appear to alter spect3-al shape. 



6. The paper by Smith and Tyler in the Journal of 

 the Optical Society of America, 5_7_, 589 (1967) v/ill be used 

 as a guide in validating the V7oods Hole experiment. 



o 



7. Our average Gray Card irradiance value at 5000 A 



3 2 



is 2.4 X 10 n watts/cm [i. Considering a diffuse reflectivity 



of the Gray Card of 0.18 v.'e obtain a downv;elling irradiance 



4 2 



value of 1.3 x- 10 [i v.'atts/cm \.i. Tyler, at Crater Lake, 



5 2 



obtained a value of 2 x 10 p v.'atts/cm [i. I would be inclined 



to accept this factor of 10 difference as partially being due 

 to differences in cloud cover (perhaps, a factor of 3) and the 

 remaining factor of 3 or 4 as a possible error in our calibration 

 due to the stacking of many neutrcil density filters. All in 

 all, we could not have hoped for nuch better results in the 

 short time period fo:'" the v;ork. I'm very encouraged v;ith 

 this compa r i'.. •:>:• i . 



29-5 



