36 



SCIKiMCE. 



[Vol. XX. No. 493 



ago I analyzed some samples of Albany water, 61tered 

 through a well-known filter manufactured in this city (the 

 Blessing Duplex Filter), and found that a sample of water 

 obtained by washing the filter after a day's use, yielded of 

 albuminoid ammonia, 0.1850 parts per 100,000, showing that 

 the filter had retained a large amount of organic matter. 

 Water which had passed through the sand of the filter only, 

 yielded 0.0023, and that which had passed through both the 

 sand and charcoal yielded but 0.0014 parts per 100,000. This 

 latter quantity is about one-tenth that ordinarily found in 

 our city water, and this is certainly a very good showing. 

 Two years since I analyzed water which had been drawn 

 from our upper service, both before and after filtration 

 through the same Altering apparatus, and found the free 

 ammonia reduced to a fifth, the albuminoid ammonia to a 

 fourth, and the oxygen absorbed to two-thirds of the amounts 

 originally present, by filtration ; while a sample of water 

 from the lower city service had its free ammonia reduced to 

 a fifth, albuminoid to a tenth, and oxygen absorbed to a 

 twelfth, indicating a vast improvement in a water at that 

 time in singularly bad condition. These results I believe to 

 be largely due to the efficient action of the animal charcoal, 

 which in this device acts, not as a strainer or filtering medium 

 proper, but as an oxidizing agent, provision being made for 

 its constant aeration. In many filtering appliances animal 

 charcoal is a fruitful source of trouble and danger, but if the 

 real filtration is accomplished by other material and the coal 

 is subjected to frequent aeration and renewed when neces- 

 sary, it is a most valuable agent for effecting the oxidation 

 of organic matter. I purpose soon to make some experiments 

 with a view to determining how long animal charcoal retains 

 its activity in such filters, though it is very certain that, with 

 proper treatment, it will continue to operate satisfactorily for 

 a long time. 



Of the chemical agents which have been employed in 

 water purification, the most important are metallic iron, solu- 

 tions of iron salts, generally the chloride, permanganate of 

 potassium, and alum. Spongy iron, obtained by the reduc- 

 tion of hematite-ore at a temperature of a little below that of 

 fusion, thereby rendering the metal porous or spongy in 

 form, was first made use of by Bischof, whose process was 

 patented in England in 1871, though Dr. Medlock had se- 

 cured a patent in 1857 for a process of purification based 

 upon the use of metallic iron plates, and Spencer in 1867 in- 

 troduced a material which he called magnetic carbide, in 

 which the active agent was iron. The eai'bonic acid in the 

 water, acting upon the iron in one or the other of these forms, 

 produces a ferrous carbonate, which, by oxidation, yields 

 hydrated ferric oxide, and this is believed to effect the oxi- 

 dation of organic matter and serve as a coagulant as well, 

 producing a flocculent precipitate, which is subsequently 

 separated by sedimentation or filtration through sand. Such 

 methods have been employed with more or less success in 

 various European cities, but Anderson's process, which has 

 been successfully used at Antwerp, Ostend, Paris, and 

 Vienna, has generally replaced other methods of purification 

 by iron. In this process the water is forced through revolv- 

 ing purifiers consisting of iron cylinders revolving on hollow 

 trunnions which serve for inlet and outlet pipes. On the 

 inner surface of the cylinders are curved ledges running 

 lengthwise, which scoop up and shower down through the 

 water fine cast-iron borings as it flows through the cylinder, 

 so that every portion of the water is brought into contact with 

 the iinn, which is kept constantly bright and clean by attri- 

 tiorj. The water issuing from the purifiers is exposed to 



the air, by allowing it to flow through a trough, to secure 

 the precipitation of the ferric hydrate, and by filtration 

 through sand this precipitate is subsequently removed. It 

 is claimed for this process that the organic matter is altered 

 in form and largely destroyed, the albuminoid ammonia 

 being reduced to from one-half to one-fourth, and micro-organ- 

 isms largely destroyed or removed. At Antwerp 2,000,000 

 gallons daily are thus treated, and Professor Edward Frank- 

 land has shown that this water is completely sterilized and 

 nearly all its organic matter removed. The cost, previous 

 to the introduction of settling reservoirs before filtration, has 

 been $4 per million gallons. In a paper read before the 

 Franklin Institute in 1890 by Easton Devonshire, C.E., it is 

 estimated that the cost of working expenses, with an output 

 of 5,000,000 gallons per diem or over, should not exceed $2 

 per million. 



Ferric chloride has been employed in Holland for removing 

 clayey matter and organic impurities from the water of the 

 Maas, which supplies Rotterdam. Carbonate of iron is formed 

 and decomposed with separation of ferric hydrate which 

 coagulates and removes the organic matter, but such treat- 

 ment is attended with many difficulties and is not likely to 

 come into general use. The same may be said of the em- 

 ployment of permanganate of potassium, which oxidizes 

 organic matter and by its decomposition yields manganic 

 hydrate which precipitates much of the suspended matter 

 present in the water. Such processes may be successful, 

 here and there, on a small scale, but they cannot as yet be 

 practically or economically employed in the purification of 

 largfe supplies. 



The only other purifying agent of which we need speak 

 is alum. It is said to have been used for centuries in China 

 and India, but particular attention was first directed to its 

 use by Jennet in 1865. Most waters contain more or less 

 bicarbonate of lime in solution, and the alum acting upon 

 this constituent yields sulphate of lime, carbonic acid gas, 

 and aluminic hydrate, as shown in the following equation: 



K2AI3 (SO4) 4+3H3 Ca (C03)3=3CaS04-f K, SO4 

 + 6 CO3+AI3 (H0)5. 



As the aluminic hydrate forms and deposits it not only en- 

 tangles and carries down finely-divided, suspended, mineral 

 matter but coagulates and removes much of the dissolved 

 organic matter as well. By this means peaty and other col- 

 ored waters are decolorized; turbid waters containing finely 

 divided clay are clarified and bacteria removed. Professor 

 A. E. Leeds, in an experiment performed upon the water 

 used at Mt. Holly, N. J., found that alum, added in the 

 proportion of half a grain to the gallon, produced the follow- 

 ing effect: " On standing the peaty matter was entirely 

 precipitated in reddish-yellow flakes and the water abo^e 

 became perfectly colorless and clear. On pipetting off some 

 of this supernatant fluid I found that instead of containing 

 8,100 colonies of bacteria per cubic centimeter, as it did before 

 precipitation with alum, it contained only 80 colonies. On 

 filtering some of this supernatant water through a double 

 thickness of sterilized filter paper into a sterilized tube I 

 found no bacteria in the filtered water. In other words the 

 water had been rendered, by the addition of an amount of 

 alum so minute as to be inappreciable to taste and almost to 

 chemical tests, as sterile as if it had been subjected to pro- 

 longed boiling." {Journal American Chemical Society, 

 ix., p. 154.) 



Austen and Wither made a valuable report to the State 

 Geologist of New Jersey in 1885, on the "Purification of 

 Drinking Water by Alum." They found that 1.2 grains per 



