August, 1907 



AMERICAN HOMES AND GARDENS 



313 



the foss is the seat of very active anaerobic fermentation, in 

 consequence of which a portion of the deposit is dissolved 

 and another portion is converted into gases, chiefly formene 

 and hydrogen, which are evolved in great quantities. Hence 

 the sediment does not accumulate and fill the foss. Dr. 

 Calmette found only about 200 cubic feet of mud in a septic 

 foss of nearly 9000 cubic feet capacity, at La Madeleine, 

 after a year of service, during which 102 tons of solid matter 

 In suspension had entered the foss. This result shows the 

 great effect of bacterial action in decomposing and dissolving 

 sediment. Part of the dissolved organic matter is also dis- 

 integrated in the septic fosses. Altogether half the organic 

 carbon is evolved in gaseous form, and much of the nitro- 

 genous matter is converted into ammonia. The water which 

 flows from the fosses is still filthy and foul-smelling, but it 

 holds no matter in suspension. 



Bacterial Contact Bed at La Madeleine-les- Lille 



In this condition it goes to the bacterial beds for further 

 purification. These beds are rectangular basins about 4 

 feet deep and from 10,000 to 20,000 square feet in area. 

 The bottom is covered with concrete, slightly inclined, and 

 has a system of tile drains. The basins are filled with coke 

 or slag broken to a diameter of one or two inches, with a 

 bottom layer, ten inches thick, of fragments from two to 

 four inches in cliameter. The water, distributed by sluices, 

 flows in little rivulets over the entire surface of the bed. 



These bacterial beds may be operated either by "double 

 contact" (intermittent system), or by percolation (contin- 

 uous system). In the intermittent system there are two sets 

 of beds, to which the water goes in succession, remaining 

 two hours in contact with each set. After the beds are 

 emptied they are left exposed to the air for four hours be- 

 fore being used again. The cycle of operations of each bed 

 comprises eight hours, thus: one hour filling, two hours 

 filled, one hour emptying and four hours airing. Thus each 

 bed is used three times in twenty-four hours. It takes a 

 volume of water equal to one-third of its cubic capacity, and 

 it is estimated that, for average conditions, a total surface of 

 five-eighths of a square foot, including the beds of first and 

 second contact, is required to purify one cubic foot of water 

 per day. Therefore five acres would suflRce for a city of 

 100,000 inhabitants with a daily sewage flow of 350,000 

 cubic feet. 



The changes which occur during these operations are very 

 complex. When the beds are flooded the slag appears to 

 extract the dissolved organic matter as fibers extract dyes 

 from their solutions. Meanwhile the disintegration of the 

 organic matter by the bacteria continues, but the deposit on 

 the slag is attacked and decomposed still more energetically 

 during the period of aeration. The products of decomposi- 

 tion are carbon dioxide water, nitrogen and nitrates, for the 



ammonia which is formed at first is oxidized and converted 

 into nitrates by certain bacteria called nitrifying ferments. 

 In this way the slag is freed from the adhering coat of or- 

 ganic matter and made ready to form a new deposit during 

 the next immersion. On leaving the first bed the water is of 

 good appearance and considerably purified, and when it 

 leaves the second bed it has parted with from seventy-five to 

 eighty-five per cent, of its organic matter. In this condition 

 it is clear, odorless and not liable to become putrid. Fish 

 thri\'e in it and it can safely be thrown into streams. 



This intermittent method has the advantage of being very 

 simple and requiring no machinery, but the output per square 

 foot of surface is still rather small. To increase it attempts 

 have been made to make the flow continuous without sup- 

 pressing the indispensable aeration of the beds. This result 

 is obtained by employing pressure sprayers, hydraulic tourni- 

 quets, inverting droppers or intermittent 

 siphons. 



The bacterial beds are made six or 

 seven feet thick, and complete purifica- 

 tion is obtained with a single bed. In 

 the first system the water falls con- 

 tinuously, in a fine rain, upon the slag 

 from spraying nozzles distributed over 

 the surface of the bed, through which it 

 percolates slowly and emerges in a puri- 

 fied condition at the bottom. The appa- 

 ratus is costly and is frequently clogged, 

 but the purification is perfect and the out- 

 put exceeds three cubic feet per square 

 foot per day. 



The hydraulic tourniquets or rotary 

 sprinklers are based on the same prin- 

 ciple. The whole surface of the bed is 



Flooding a Bacterial Bed of First Contact at La Madeleine-les- Lille 

 Part of Second Contact Bed Is Shown at Right 



sprinkled by a single central apparatus with two or four 

 hollow and perforated arms. In this system also the purifi- 

 cation is perfect, and the output is nearly four cubic feet per 

 square foot per day. But here, too, the apparatus is ex- 

 pensive and liable to stoppages. Besides, it is often disturbed 

 by the wind. 



The inverting droppers usually consist of a wheel carrying 

 buckets which empty themselves on the bed. The wheel 

 automatically moves forward at each emptying and so the 

 entire bed is watered. One of the best devices of this class 

 is Liddian's rotary distributor, which gives results as good 

 as those obtained with sprinklers and works with much 

 greater regularity. 



Finally, the intermittent siphons devised by Dr. Calmette 

 are simply siphons which fill automatically at the expiration 



