Introduction 



This volume presents tabular data and certain charts and graphs in the general field 

 of nutrition and metabolism. Both plant and animal forms are included. The guiding prin- 

 ciple in selecting material has been that it be of basic importance in its general field; ready 

 availability elsewhere has not been regarded as a reason for excluding it. 



Some material of fundamental importance has had to be omitted, either because 

 efforts to secure the needed data have not succeeded, or because time has not permitted 

 the necessary steps for getting available data into print. Inability to publish material that 

 a contributor may have spent many hours in compiling is cause for the deepest regret. 

 Such unused material will, it is hoped, be the source of valuable additional tables when this 

 volume is revised. 



In the preparation of material in tabular form the chief objective has been clarity 

 of presentation. Where the subject matter of a table has been considered to be inherently 

 difficult for the non-specialist -- the beginning student in the subject or the specialist in 

 another field -- effort has been made, in explanatory headnotes and footnotes, to resolve 

 for the reader some of the difficulty. Footnotes have also been used in many instances 

 for information originally within the table itself where simplification of the structure of the 

 table could be achieved by withdrawal of the information into the footnotes. 



In each instance where a numerical value is given for a variable, the value is the 

 mean (or adjusted mean) of a group of measured values. Where possible to obtain, each 

 such value is followed by an estimate of the lower and upper limits of the 95% range. The 

 95% range has been selected in preference to the standard deviation as better suited to the 

 needs of the reader who is not a specialist in the field from which a value has been drawn. 

 The 95% range is a direct representation of the ordinary* range of variation, to be had only 

 by further calculation if the standard deviation alone is available. The latter has the dis- 

 advantage of not being readily available in 

 many instances and of giving biased limits 

 for the 95% range when a variable has a 

 skewed distribution. The statistically- 

 minded reader who might wish to make fur- 

 ther calculations from values in these tables 

 will not care to proceed without information 

 on comparability and number of measure- 

 ments. Unfortunately space does not per- 

 mit including here such collatereil informa- 

 tion,, but the bibliographic references will 

 lead to the original data where it should be 

 found. 



The 95% range may be estimated 

 in several ways, the method depending 

 on the information available. The types 

 of estimate most commonly encountered 

 are listed below. The letter designations 

 (a, b,c,d) will be found as identifying 

 superscripts opposite ranges given in the 

 tables. (For details of these and other 

 estimates see texts on statistical methods.) 



95% Range- 



*To the clinician, equivalent, with reservations, to "normal and borderline" 



