74 ANDREWS. 



mother's diet we found that it consisted entirely of white rice, 

 and fish or meat; rarely were any vegetables or fruit eaten. 



The disease is just as apt to manifest itself in the first infant 

 of a girl 18 years old as in the infant of an older woman with 

 her third or later pregnancy. Indeed one of the severest cases 

 we saw in both mother and child was exhibited in a girl barely 

 18. The child died when 6 weeks old and at this time the mother 

 could hardly stand. She sat on the floor and pushed herself 

 along as best she could. The pains in the legs were severe. 

 Areas of anaesthesia, numbness, and formication were present. 



It is not unusual to obtain the history that the first and second 

 children have died each at about 2 months of age from taon. 

 With the third child the doctor advises artificial feeding, and if 

 the little one escapes the gastro-enteritis which usually follows, 

 it survives. With the fourth child the mother resumes the 

 breast-feeding with the result that the infant dies of taon in about 

 2 months. Occasionally the mother starts the infant on artificial 

 feeding. In about a month, gastro-enteritis has become severe, 

 and she returns to breast feeding; two months later the child 

 dies of beriberi. 



Several families were found in which the first and second 

 children were born in the country and showed no evidence of 

 taon. On moving to Manila the subsequent children were car- 

 ried off by this disease. In the country the daily rice is pounded 

 out by the family and a large part of the pericarp remains. 

 Furthermore, both vegetables and fruit form a part of the diet, 

 whereas in Manila only polished rice is obtainable. 



Analysis of milk. — In the past attempts have been made to 

 analyze the milk of women with beriberi. In all such attempts 

 the diagnoses were based only on clinical findings in the child or 

 mother or both. The amount of milk secured was small and 

 the number of cases very limited. It is evident that analyses 

 under such conditions are unsatisfactory and perhaps misleading. 

 In the present cases the milk was obtained from women the 

 death of whose infants from beriberi had been confirmed by 

 necropsy. Results obtained from such cases are obviously more 

 reliable than those obtained from cases based only on clinical 

 symptoms; then, too, the amount of milk secured was larger. 

 The mother was visited as soon after the funeral of the child as 

 possible. She was given a sterile flask containing a few drops 

 of formalin and was told to milk into the flask all she could 

 from time to time. The next day at about the same hour the 

 flask was called for. In this way we secured a twenty-four-hour 



