NUTRITIVE VALUE OF RAW AND BOILED MILK 199 



was also worked out. In this latter case the calculations were 

 based on the weights of each series on the sixteenth day, the data 

 of earlier periods being regarded as unreliable. The percentages 

 given by the first method have been charted in relation to the age of 

 the infant, and are shown in Diagram 9 on the preceding page. This 

 shows that the actual percentage increase in weight from one eight- 

 day period to another, of all the infants of each series considered in 

 the aggregate, is extremely variable ; even where infants of precisely 

 the same age are considered in large numbers, as in this investigation, 

 there appears to be no constant rate of growth from birth onwards. 

 The figures showing the rate of growth are given in Appendix D. 

 The percentages, although (except in two cases) positive, show an 

 extraordinary variation, and differ in range between 0*6 and 

 0-8 per cent, in the breast-fed series, and between 1-4 and 6'8 in 

 the artificially fed series, during the life-period now under con- 

 sideration. It is frequently believed by those concerned with the 

 feeding of infants that the infant's weight should show a constant 

 increase from day to day. Doubtless this does occur in a con- 

 siderable number of children, although in most infants the amount 

 of increase will vary very considerably not only from day to day 

 but from week to week. The present investigations, however, 

 seem to indicate sufficiently clearly that a large number of children 

 who are reasonably healthy show much variety in relation to their 

 progress from day to day. In many cases a slight loss occurs, to 

 be followed at a succeeding period by a larger rise than that of the 

 loss incurred. 



The fluctuations appear to be slightly less marked in the arti- 

 ficially fed infants than in the breast-fed infants. In both series 

 there is a tendency for the percentage increase between any two 

 consecutive periods to become less as the age advances. 



It has long been recognised that the infant gains weight more 

 rapidly in the first few months of life than it does later, the progress 

 becoming slower after the first few months. 



If, now, the results of the second method, showing the rate 

 of increase of weight of the infants in the two series from the 

 sixteenth day of life onwards, be considered, it is found that from 

 the beginning of the period up to about the fifth month there 

 is very little difference in the curves. This absence of 

 difference is accounted for by the fact that the artificially-fed 

 infants show for the most part a lower absolute weight. Hence 

 a less increment is needed in order to show a similar percentage 

 increase to that of the breast-fed series. This is shown on the 

 diagram on p. 200. It may be observed that after approximately 

 the fifth month, the artificially-fed infants gain at a slightly 

 greater rate than the breast-fed infants. For figures see 

 Appendix D. 



The deficit in weight of the series of artificially-fed infants 

 over that of the breast-fed infants was also investigated, and 



