to determine with certainty whether any of the bacteria found in the effluent came down through the filtering material with the applied water. The sanitary importance of determining this point is obvious, since one of the principal objects of water filtration is the removal of pathogenic bacteria. Elaborate preparations were therefore made to settle this difficult question. The degree of bacterial purification accomplished by Filter Tank No. 8, indicated by the number of bacteria in the water before and after passing through it, is shown by the following table : Table of Monthly Averages of the Number of Bacteria found in a Cubic Centimeter of City Water and in the corresponding Effluent from Filter Tank No. 8. In January, 1890 (see Special Report on Purification of Sewage and Water, 1890, page 615), it was shown that there was a growth of bacteria in the underdrains and outlet pipe of this filter. This was proved by examination of the water expressed from a sterilized sponge after wiping out the pipe. This operation was repeated several times in 1891. The species present in the air and dust were obtained by exposing a sterile gelatine plate in the vicinity of the outlet pipe. These species of bacteria, as well as those found in the outlet pipe and underdrains, were quantitatively determined. In the following table a comparison is shown of the results of the quantitative determination of the species of bacteria in the applied city water, in the effluent, the air, the outlet pipe and the underdrains. In the case of air the figures indicate the percentage which each species forms of the total number of bacteria on the plates. It is to be noted that it was during this period that the quantitative method of species determinations was developed, and, although the work was done as carefully as possible, the results are somewhat less accurate than those obtained after more experience with the method. Average Results of Quantitative Determinations of the Species of Bacteria. May to August, inclusive, 1891. SPECIES OF BACTERIA. (AVERAGE NUMBER PER CUBIC CENTIMETER.) It will be seen that the species are designated by numbers, and not by names. For reasons stated above, this work has been one of comparative and quantitative species determination, rather than a detailed study of the species themselves. As nearly as can be learned, however, from the comparisons made up to this time, the numbers used in these tables signify, and doubtless in most cases are identical with, previously described species, as indicated below:No. 26 B. cereus, Frankland = In studying the first portion of the table given above, it will be seen that species Nos. 20 and 11.0, which form more than eighty per cent. of the bacteria in the applied water, constitute less than forty per cent. of those in the effluent. These two species are very prominent ones in the underdrains (No. 20) and in the air (No. 11.0). A comparison of the remaining species indicates, also, that it is far more reasonable to conclude that they have had their origin in either the underdrains or the air, than that they have passed with the applied water (containing very small numbers) through the filtering material. The second portion of the table shows a very striking resemblance between not only the species of bacteria but their relative numbers, in the effluent and those in the underdrains and outlet pipe. The relation of the species and their numbers, in the case of the bacteria of the applied water and the effluent, is far less marked; while two species present in the applied water, No. 11.2 = B. coli communis, and No. 55 B. aurantiacus, were not found in the effluent. The investigation up to this point confirmed the results of the work of previous years; namely, that the few bacteria in the effluent have their origin in the underdrains and outlet pipe, and do not pass through the filter. In order to obtain still more definite data, it was determined to increase as much as possible the number of applied bacteria, and to guard, with the utmost care, against contamination of the effluent. To this end canal" water taken from the Merrimack River above Lawrence, and containing from one thousand to fifteen thousand bacteria in a cubic centimeter, was applied to the filter, commencing Nov. 13, 1891. The effluent was protected from contamination by dust and air by covering the outlet pipe with a large zinc-lined box, which was frequently washed with corrosive sublimate, and so arranged that the stopper of a sample bottle could be removed beneath it and all manipulations could be done without disturbing it. From the date of application of canal water to the filter several determinations of the number of bacteria in the applied water and in the effluent were made daily from samples collected at different hours. The results are as follows: Summary of Bacterial Counts. (Nov. 13-Dec. 31, 1891.) Of the one hundred and two determinations made of the effluent, fifty-eight showed no colonies on the gelatine plates. The determinations of the number of bacteria in canal water and in the effluent from Filter Tank No. 8 were made on the same gelatine, grown for the same length of time and at the same temperature. Many duplicate determinations were made of the effluent, and the plates were allowed to stand eight and even ten days before counting. They agreed very well with those counted after three days; rarely a colony was found which proved to be a spore-former, and had no relation to the leading species in the applied water. A quantitative species determination was made of all samples of Filter Tank No. 8 effluent, as well as of canal water, at frequent intervals. The results are given below. It will be seen that these determinations indicated the presence of twelve species in the canal water. This does not mean that they were the only species of bacteria present, but that these species were the leading ones, occurring in numbers sufficiently great to be found on plates of samples diluted one to one hundred. Number of Bacteria of Each Species found in a Cubic Centimeter of Canal Water. |