8 BULLETIN 36, U. S. DEPARTMENT OF AGRICULTURE. 
in price less than $2.75, while 3 similar samplings in Norman de- 
veloped differences of only 75 cents among 9 middling bales on Octo- 
ber 30. $1 among 7 middling bales on October 25. and a range of 
$2.50 among 11 middling bales on November 4. It is notable that 
the narrowest range of prices shown on middling cotton in any one 
town on any one day is 62 cents at Clinton on October 26, and this 
is probably explained by the fact that all 4 of the middling bales 
sampled in Clinton on that date were purchased by the same buyer. 
In all the other figures presented in this table great care has been 
taken to avoid including any bales which were handled as parts of 
round lots. Of course we have no means of knowing that in every 
case all of the bales which we have found to grade middling were so 
graded in the first transaction; in fact, the wide range in prices 
would seem to indicate that the price on a great many middling 
bales must have been seriously depressed for some reason which our 
critical examination of the samples failed to develop. 
This discussion of the variation in the price of middling cotton 
might be continued and emphasized by numerous citations from col- 
lections which include smaller numbers of middling bales. Three 
examples will suffice to show that these irregularities are even more 
widespread than Table I shows. For instance, the collection of 
samples from about a dozen bales which had been accumulated in a 
dealer's yard at Lone Wolf, Kiowa County, and which were pur- 
chased on almost as many different dates, showed a difference of 
$2.50 in the prices paid for 2 bales of middling cotton on the same 
date. Of a similar lot sampled at Cordell, in Washita Comity, which 
had been purchased just one day later than the bales at Lone Wolf, 
a difference of $5 per bale was found. A collection from the street 
sales at Marietta on December 4, when most of the cotton was low 
grade and only 2 bales of middling were found, showed that one 
grower received $5 more than the other for his middling bale. When 
such variations in the price of middling cotton are developed by the 
random sampling of 2 or 3 bales of that grade in a town, it is almost 
appalling to consider the injustices which must be done in the aggre- 
gate in transactions involving only the best-known grade of cotton 
and that upon which the daily quotations are based and the daily 
purchasing limits set. 
An examination of the results of this investigation indicates that 
somewhat greater irregularities occur in the handling of the higher 
and lower grades than those shown for middling cotton, the most 
notable being in the very low grades, as shown by Tables II. Ill, IV, 
and V. The fact that the price of middling is often more variable 
than any other is brought out by a further analysis of the figures 
for the samples taken at Cordell from the bales sold on November 8. 
A total of 13 bales were sampled. Of these. 3 were middling, and the 
