58 DISCUSSION 
Fic. 11—Illustrations of sky code, tropical disturbance to typhoon 
Typical illustrations of each sky type (and 
most of the postscript modifications thereof) are 
shown in Figs. 9-11. These were chosen from all 
the prints of the flight films for a combination 
of best representativeness and reproducible pho- 
tographie quality. 
Discussion 
Dr. Malkus—I think this is an excellent il- 
lustration of how a picture can say much more 
than many words. It was with considerable sur- 
prise that we did not find the wave again after 
it had passed over us during the night and was 
presumably a hundred miles beyond Kwajalein. 
It simply was not there anymore. This illustrates, 
as we saw in Figure 8, that in order to make 
an analysis of precipitating systems in the 
tropics, we not only have to know about the 
synoptic patterns at all levels, but also about the 
intense coupling between the high tropic patterns 
and those in the low troposphere. 
I will open the discussion by making one 
comment. It is a provoking question; namely, 
how well could we predict, understand, or even 
regulate the microphysics of tropical rain, if 
we were in an aircraft in the Marshall Islands 
area and, out of context, made a microstudy of 
the clouds? Other aircraft studies might be made 
in the area of Kwajalein and Majuro, which was 
on the other side of the wave. The result would 
probably be that one group of people says tropi- 
cal clouds are less than 10,000 ft high and 
totally disorganized, while the other group says 
the observations are wrong, the clouds extend to 
50,000 ft and are well organized. I have not 
heard any arguments quite that preposterous, 
but, of course, in the different sections and under 
the different conditions under which these ob- 
servations were made, we would arrive at dif- 
ferent conclusions. 
Dr. Horace R. Byers—I think that the great- 
est sin that cloud physicists commit is to draw 
conclusions from small samples. 
Dr. Morris Neiburger—Well, I will not say 
that what Drs. Malkus and Byers have just 
said may not be true, but it seems to me that 
the case is a rather different one. Rather than 
assuming that all clouds have the same micro- 
