6 CM. JACKSON 



(cavities opened and blood-clots removed) ; lungs (right and left 

 separately); liver; spleen; stomach and intestines, including con- 

 tents, mesentery and pancreas; same, without contents; supra- 

 renal glands ; kidneys ; ovaries or testes (including epididymis) ; 

 spinal cord (in a few cases). The extremities, skin, skeleton and 

 musculature were also weighed in some cases, as described in a 

 separate paper by Jackson and Lowrey ('12). 



The organs were weighed in closed containers, and loss by 

 evaporation was avoided so far as possible. The organs were 

 usually weighed to 0.1 mgm. (0.0001 gm.) excepting some of the 

 larger organs in the older rats, which were weighed to 1 mgm. 

 The observations were recorded on printed cards, and any unusual 

 conditions carefully noted. The calculations were made by the 

 aid of a Burroughs adding machine, Crelle's Rechentafeln, and 

 the tables in Davenport's 'Statistical Methods.' The calculations 

 were carefully checked independently to eliminate errors. 



I am greatly indebted to Dr. S. Hatai of The Wistar Institute, 

 for valuable aid and criticism, especially on the mathematical 

 phases of the work. 



In calculating the various .statistical constants — mean, stand- 

 ard deviation, coefficients of variation and of correlation and 

 probable errors — the usual formulas (Pearson's) were employed 

 as given by Davenport ('04). The ungrouped data were used in 

 all cases. 



Since the present paper is concerned largely with variability 

 as measured by the coefficient of variation, a brief discussion of 

 this coefficient may be desii'able. The coefficient of variation is 

 one hundred times the ratio which the standard deviation bears 

 to the mean; or the percentage of the mean which the standard 

 deviation forms, as expressed by the formula: 



C = ^100(%) 

 For the standard deviation, the formula is as follows: 



(X = 



n 



