X, A, 2 Heise: Water Supplies in the Philippines 137 



by the analytical records of the Bureau of Science, where waters 

 known to be uncontaminated range as high as 20 or more parts 

 per million in free ammonia content. Therefore it seems desir- 

 able to eliminate nitrogen determinations in most cases and to 

 limit our efforts to the determination of normal mineral con- 

 stituents, supplemented by biological examinations whenever 

 feasible, until such time as our increased knowledge of Philippine 

 waters and of tropical conditions shall justify other methods. 



The analytical data obtained in the Bureau of Science labora- 

 tory during 1914 are contained in Tables III to V at the end of 

 this paper. All results are given in terms of parts per million 

 unless otherwise stated. 



FIELD WORK 



It soon became evident that a field survey offered the best 

 solution of the problem of obtaining the most information con- 

 cerning Philippine water supplies with the least possible expend- 

 iture of time and money. Accordingly a simple chemical field 

 outfit, based on that described by Leighton,^ was constructed. 

 Field methods leave something to be desired, so far as accuracy 

 is concerned, but it is thought that they have been more than 

 justified by the volume of reasonably accurate work done in a 

 comparatively short time, and by the intimate knowledge of local 

 conditions made possible by the field trips. 



In addition to the chemical equipment, the field outfit included 

 a small portable bacteriological kit, consisting of Petri dishes, 

 a case of sterile pipettes, and tubes of litmus lactose agar and 

 lactose bile agar. A plate culture on litmus lactose agar gave 

 an approximate bacteria count and showed the presence or 

 absence of acid-forming bacteria; a tube culture in lactose bile 

 agar indicated the presence or absence of lactose-fermenting 

 organisms. The uniformly high temperature encountered in 

 most parts of the Islands enables fairly concordant results to 

 be obtained without the use of an incubator. Pipettes were 

 separately wrapped with gauze and carried in a tinned cylinder ; 

 hence it was easy to sterilize them even in the field. The 

 Petri dishes were simply wrapped in paper, in packages of six, 

 before being sterilized ; they remained uncontaminated for 

 weeks. Bacteria counts were generally made at the end of both 

 twenty-four- and forty-eight-hour intervals. 



The field investigations were, in the main, confined to three 

 islands: namely, Mindoro, Cebu, and Panay. The analytical 



■'' U. S. Geol. Surv., Water Supj^ly Paper (1905), 151, 



