POSTNATAL GROWTH IN THE ALBINO RAT 21 



true correlation being augmented by a 'spurious correlation,' as 

 will be explained later. There appear no constant differences 

 according to age and sex. If such exist, as is probably the case, 

 they are obscured by accidental variations in the comparatively 

 small number of observations. 



The close correlation between the weights of the head and of 

 the whole body is also evident from the slight variability in the 

 percentage weight which the head forms of the whole body (table 

 3). If the head at each age formed a constant per cent of the 

 body, the variabihty of the percentage weight would be zero. 

 The coefficient of variation of the percentage weight of the head 

 is, however, usually not very much smaller than that of the 

 absolute weight, and in a few cases even appears slightly larger. 

 That is to say, at any given age the absolute weight of the head 

 can be predicted almost as accurately as the percentage weight. 

 This is not the case with most of the individual organs, as will 

 be seen later. 



Whether reckoned upon the absolute or upon the percentage 

 basis, it is evident the head of the rat exhibits slight variability 

 when compared with most other organs as well as with the body 

 as a whole. This agrees with the observations of Hatai ('08) 

 upon the variability of the skull bones of the rat, as well as with 

 what has been found in the human body. Quetelet ('71) and 

 other anthropometrists have observed that the head is the least 

 variable portion of the human body, a conclusion supported by 

 extensive and careful measurements on the skeleton of different 

 parts of the body by various observers. 



2. Central nervous system 



Since the growth of the brain and spinal cord has been care- 

 fully worked out by Donaldson ('08) from extensive data, it was 

 thought unnecessary to include these organs in the present inves- 

 tigation. Incidentally, however, some observations were made 

 upon the brain, and a few upon the spinal cord. From these, 

 together with those published by Donaldson ('03) and some 

 unpublished data from The Wistar Institute, a few additional 

 conclusions may be drawn. 



