Amsepctur'i92hiarm'  j       The  Relat^ve  Value  of  Protein*.  647 
with  wheat  gliadin ;  this  is  a»  complete  protein,  but  has  very  little 
lysine.  Adult  rats  were  maintained  for  long  periods  on  this  pro- 
tein— as  long  as  500  days — but  young  rats,  though  they  lived  for 
long  periods,  failed  to  grow.  The  authors  therefore  regarded  lysine 
as  essential  for  growth.  Later,  they  showed  that  if  they  added  lysine 
at  definite  intervals  to  the  food  containing  gliadin  as  sole  protein, 
growth  took  place  with  the  lysine,  but  not  without  it.  The  mini- 
mum amount  of  lysine  necessary  to  produce  normal  growth  was 
found  to  be  between  2  and  3  per  cent,  of  the  amount  of  protein 
in  the  diet. 
The  value  of  lysine  for  growth  was  shown  in  a  more  practical 
way  by  Buckner,  Nollau  and  Kastle.  They  fed  chickens  on  a  poultry 
farm  on  diets  of  high  and  low  lysine  content.  Their  figures  and 
photographs  definitely  showed  more  rapid  growth  on  the  mixture 
of  high  lysine  content. 
The  two  other  hexone  bases — arginine  and  histidine — are  essen- 
tial units  of  proteins.  As  shown  by  Ackroyd  and  Hopkins,  their 
removal  from  the  hydrolysed  casein  mixture  leads  to  loss  of  weight 
of  the  animals;  if  only  one  is  absent  from  the  food  the  rate  of 
growth  is  lessened,  so  that  it  appears  as  if  these  two  units  were 
inter-related.  The  structural  formulae  of  these  compounds  suggest 
a  possibility  of  the  conversion  of  the  one  into  the  other  by  the  won- 
derful mechanisms  of  the  animal  body.  These  two  units  were  found 
to  give  an  increase  of  allantoin  in  the  urine  of  the  animals ;  thus 
we  know  that  the  purine  ring  can  be  synthesised  from  them.  This 
synthesis  by  animals  has  long  been  suspected,  as  young  birds  and 
animals  produce  purines  on  a  diet  with  an  almost  complete  absence 
of  purine  compounds. 
It  is  more  difficult  to  arrive  at  the  function  of  the  amino-acids 
containing  aromatic  nuclei.  Some  proteins  lack,  tyrosine,  but  all 
contain  phenylalanine,  which  is  very  difficult  to  remove  from  a 
mixture.  Phenylalanine  and  tyrosine  both  give  rise  to  homogen- 
tisic  a,cid  in  cases  of  alkaptonuria;  phenylalanine  may  therefore  be 
oxidisable  to  tyrosine  in  the  body.  The  almost  complete  removal  of 
tyrosine  from  the  mixture  from  casein  made  no  difference  to  the 
growth  of  rats,  as  shown  by  Totani.  The  amount  of  phenylalanine 
in  the  mixture  was  thus  probably  sufficient  to  supply  the  need  for 
aromatic  nuclei.  According  to  Abderhalden,  tyrosine  cannot  be  dis- 
pensed with. 
