10 



TECHNICAL BULLETIN 17 



Spores from these various sources were then measured for length. 

 In all cases, measurements were made with a Bausch and Lomb micro- 

 scope, using the 4 mm. objective and an eyepiece micrometer calibrated 

 so that one space was equal to 3.4 microns. It was observed that at 

 an extreme temperature, such as 32 C, there was a great deal of 

 variation in length and a large number of measurements would be re- 

 quired to obtain a normal curve. Data were therefore recorded for 

 the measurements of 100, 300, 500, and 1000 spores. From these data 

 the mean was calculated for each group and the differences were com- 

 pared in relation to the probable errors, according to the methods given 

 by Babcock and Clausen (2). These data are summarized in Tables I 

 and II. For 100 spores the mean was found to be 62.30 0.85 microns ; 

 for 300 spores 57.44 0.57 microns; for 500 spores 59.66 0.45 

 microns; and for 1000 spores 59.07 0.31 microns. Their accuracy 

 can be seen by comparing these results with their probable errors. In 

 Table II the comparison of the means for 100 and 300 spores with 

 the means of each of the other three groups, shows that the difference 

 between any two is from 3 to 5 times the probable errors of the differ- 

 ence. This borders on the verge of a significant difference, so that 

 100 or perhaps even 300 spores are scarcely enough to use as a basis 

 for drawing conclusions. When the mean for 500 spores is compared 

 with that for 1000, the ratio is i :i. The results obtained by measuring 

 1000 spores are only very slightly more accurate than those obtained 

 by measuring 500 spores. The difference is certainly not great enough 

 to necessitate the measurement of the second 500 spores. 



TABLE II 



SUMMARY OF COMPARISONS BETWEEN MEANS AND COEFFICIENTS OF VARIABILITY FOR LENGTH OF 



SPORES OF Helminthosporium sativum OBTAINED FROM MEASURING POPULATIONS 



OF DIFFERENT SIZE (FROM DATA SUMMARIZED IN TABLE I) 



