VITAMIN A 267 



body weight, in part in the larger animal to a less-pronounced impulse 

 to grow, and perhaps in part to a shortage of some growth factor still 

 unrecognized, and not so abundantly provided or stored as to permit 

 indefinite uniform growth in these vitamin A experiments. 



The probability that mammalian nutrition requires more than the 

 six well-recognized vitamins is discussed elsewhere in this monograph. 

 In their discussion of vitamin A testing, Drummond and Morton 

 (1929) say: "In any case the test must not be too prolonged, for even 

 on doses producing normal or nearly normal growth in the early 

 stages for a time, the response may later show a failure which is not 

 made good by increasing the amount of the supplement." These authors 

 consider it reasonably safe to use a test period of 5 or 6 weeks (after 

 depletion). 



Vitamin A in Plant and Animal Tissues and Products 



The distribution of vitamin A in nature has been the subject of 

 numerous researches in the past few years. As explained in the fore- 

 going section, its determination has become quantitative in character and 

 relative amounts of the factor may be expressed in terms of recognized 

 standards or "units." This has a decided advantage over the former 

 classification of foods as "rich,'' "poor," or "intermediate" sources of 

 the vitamin in that it permits of finer distinctions and of more concrete 

 comparisons and applications. 



It is now known that in much of the early work, vitamin A has 

 not been the only limiting factor in the growth of the experimental 

 animal. Furthermore, certain important refinements of the method 

 have not been employed in all cases even of recent work. But granting 

 these facts, our knowledge of the occurrence and estimation of vitamin 

 A is clearly passing out of the qualitative stage and hereafter may be 

 organized in a more exact form. At the present time, a comprehensive 

 survey of our knowledge of the natural distribution of vitamin A will 

 involve the use of both qualitative and quantitative data. 



Since the early experiments of McCollum and Davis (1913) and 

 of Osborne and Mendel (1913a, 1914) showed that certain foods such 

 as butterfat, egg fat, cod-liver oil and the fat of pig's kidney are rich 

 sources of the fat-soluble factor and that lard and vegetable fats in 

 general seem to lack it, the data have been extended over a very wide 

 range of foods. 



The fact that they are so numerous makes it undesirable to include 

 all of them here ; but typical foods from a wide biological range of 

 sources mav be mentioned. 



