Page images


There is a characteristic microscopic flora of soil, which is different from that of any other natural medium. The following groups of micro-organisms develop on aerobic gelatin plates inoculated with soil:

I. From 5 to 10 per ct. spore-formers (the B. subtilis group). Nearly all the colonies of these bacteria, however, seem to come from spores instead of from active organisms.

II. Under 10 per ct. rapidly liquefying, non-spore-forming, short rods with polar flagella (principally Ps. fluorescens).

III. From 40 to 75 per ct. slowly liquefying or non-liquefying, nonspore-forming, short rods.

IV. A few micrococci. In cultural characteristics these are almost identical with the last mentioned group.

V. From 12 to 50 per ct. Actinomycetes.

Of these five groups, the most important ones seem to be Nos. III and V. This conclusion is based upon the following facts: These two groups are always present in large numbers; of the other three groups, the only one always present in large enough numbers to be detected on the plates is No. 1, the group of spore-formers, which apparently exists in normal soil only in the form of spores.





The technic for soil bacteriology has never been standardized. In dairy bacteriology certain official methods are recognized, which are quite rigid even in small details; but there have been no serious attempts to propose official methods for soil study. The reason for this difference is partly because soil methods are still in their experimental stage, and partly because no extensive routine examinations of soil are ever made. If such examinations were made, they would create a demand for the general use of methods that could be counted on for comparable results. It is really fortunate that no official methods are needed, as each investigator in soil bacteriology is free to develop his own technic; but as a result the investigator must always publish a detailed description of the methods he uses. The present paper is a discussion of the methods used in these Soil Flora Studies, the reasons for using them, and the lines along which these methods are likely to develop in the future.


The exact method of taking the soil samples has not been considered a matter of great importance. It was ordinarily unnecessary to secure a representative sample, as the object of the work was to find as many kinds of bacteria as possible rather than to make a quantitative comparison between the different soils. Nevertheless care was generally taken to make borings at different spots in the plat sampled and to mix thoroly all the soil obtained from these borings. Samples of field soil were generally taken with an auger to the depth of six inches. Pots were sampled with a cheese-sampler long enough to extend to the bottom of the pot; and a little soil was carefully selected from along the entire length of the core thus extracted. It was considered unwise to make more than one boring in a pot at each sampling, because of the amount of aeration thus produced.

Sterilization of the sampling tools was also considered unnecessary. They were always well cleaned, but not sterilized. With the great dilutions used, there is no reason for supposing that contaminations due to unsterilized sampling tools would ever cause the occurrence of more than one or two colonies per plate; and no type of organism has been considered in this work unless it has been of frequent occurrence and has appeared on the plates in more than isolated colonies. It was soon found, moreover, that the soil flora was so constant that any gross contamination could be easily detected by means of the presence of unusual types of colonies on the plates.


General technic.— The only satisfactory method yet devised for determining the numbers of bacteria in soil is by means of poured plates. It has long been recognized that this method does not give a complete count. There are bacteria in soil that do not grow in our ordinary culture media, and as this method gives a count of those organisms alone that are able to produce macroscopic colonies in agar or gelatin, those that do not grow are overlooked. Moreover, if the bacteria in soil occur in clumps that do not break up while the material is being prepared for plating, each clump produces but one colony, no matter how many individuals it may contain. Still further errors are introduced into counts made this way by the fact that differences in technic cause variations in the number of bacteria able to produce colonies.

This last source of error was taken up at some length in Technical Bulletin No. 38.35 At that time the importance of a long incubation at a low temperature was emphasized. Gelatin plates, it was stated, should be incubated at temperatures as low as 18° C. for at least seven days, while agar plates should be held at least as long if not longer. Considerably lower counts were obtained with shorter periods of incubation.

Since that paper was published, some further work has been done to see what sort of incubation is best for agar. As agar can be kept at a higher temperature than gelatin, it seemed possible that more satisfactory results might be obtained in a warmer incubator. The results of this test are given in Table I. The two temperatures used


(Soil used was Dunkirk Silty Clay loam, kept in the laboratory in pots.)


Medium used.

Colonies developing per gram of dry soil with


Incubation at 18° C.

Incubation at 25° C.



agar with:

[blocks in formation]



[ocr errors]
[ocr errors]

1.... Nov. 15 1% glycerin. 26,000,000 *30,000,000 23,500,000 23,500,000 2. . Nov. 15 1% glycerin. 15,500,000 19,500,000 23,500,000 25,500,000 3.... Nov. 221% glycerin.

5,000,000 6,700,000 7,000,000 4. Nov. 23 0.5% glycerin... 63,000,000 76,000,000 68,000,000 72,000,000 5 Nov. 23° 0.5% glycerin.. 12,000,000 14,500,000 19,000,000 19,000,000 6. Dec. 21% glycerin. 16,000,000 25,000,000 28,000,000 30,000,000

Dec. 2] 1% glycerin. 10,500,000 16,000,000 16,500,000 17,000,000 8. Nov. 301 1% glycerin.

12,000,000 14,000,000 14,000,000 9. Nov. 301 0.5% glycerin.

[ocr errors]


9,500,000 14,500,000 15,000,000

* The higher 14-day count in each test is printed in bold-faced type.

35 See footnote 22.

were 18° C. and 25° C. A slight advantage seems to be indicated for the 25° incubation, as in all but two of the nine cases the higher count was obtained at the higher temperature. Generally the difference in favor of the higher temperature has been slight, although occasionally (as in test 2) it was of considerable size. It is of especial interest to notice that there has seldom been an appreciable increase in the count during the second week at 25°, which means that if 25° is used, a seven-day incubation is sufficient. In the present work, however, 18° for 14 days was used regularly, because it proved more convenient to use the same temperature for both gelatin and agar, and the differences between the two temperatures were so slight as to be of little consequence. There seems to be no good reason for recommending any special temperature for agar plates, provided it is not too high, and the period of incubation is long enough to allow all the colonies possible to develop.

Another point that proved to be of considerable importance in its influence upon the count was the number of colonies allowed per plate. When using gelatin, preference was given to counts obtained from plates having between 50 and 150 colonies each, as with smaller numbers chance variations would have too great an influence on the count, and with greater numbers the count was too likely to be reduced because of liquefaction or overcrowding. Gelatin plates with as many as 200 or as few as 30 colonies were sometimes used when others were not available, and in many cases they were probably as reliable as those within the narrower limits. Agar plates apparently give reliable counts with as many as 300 colonies per plate. Ordinarily, however, the number of bacteria in soil is so nearly constant that it is easily possible to obtain plates with more than 50 and less than 150 colonies each. The dilutions 1 to 100,000 and 1 to 200,000 were ordinarily used. These two dilutions gave the desired number of colonies per plate, unless the count was under 5,000,000 or over 30,000,000 per gram of moist soil — which rarely happens in field soils. In some special cases in which abnormally high numbers were expected, such as when the soil was frozen, recently aerated or manured, it was necessary to use greater dilutions.

Technical Bulletin No. 38 dealt mainly with the subject of media. It was shown that various media gave very similar counts. If anything, the highest counts were obtained upon a soil-extract gelatin; but an asparaginate agar 36 was especially recommended, , principally because of its definite chemical composition. Since that paper was published, further modifications of both these media have been made.

36 Distilled water 1000 C. C., agar 12 g., dextrose 1 g., sodium asparaginate 1 g., NH,H,PO, 1.5 g., CaCl2 0.1 g., MgSO. 0.2 g., KCI 0.1 g., FeCl3 trace. Reaction adjusted to 1.0 per ct. normal acid to phenolphthalein.

Comparison between soil-e.ctract gelatin and tap-water gelatin.--A few figures given in the paper just mentioned indicated that tap-water might be substituted for soil-extract in the gelatin; but at that time so few results were available that it hardly seemed wise to recommend tap-water gelatin. More recently eleven more tests comparing soil-extract gelatin and tap-water gelatin have been made (Table II). The counts upon the two media were generally almost the same; and the counts on soil-extract gelatin were higher than on tap-water gelatin in only three of the eleven tests. There seems, therefore, to be no reason for preferring the soil-extract formula; and as tapwater gelatin is of simpler composition, it is now recommended.


[graphic][merged small][merged small][subsumed][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small]

The higher count in each test is printed in bold-faced type.

Comparison between different brands of gelatin.-- A footnote in Technical Bulletin No. 38 stated that the use of 20 per ct. gelatin, instead of 12 per ct., prevented rapid liquefaction and thus allowed a higher count. Some work has been done since then to determine what concentration of gelatin is best. Besides Gold Label gelatin, which was used thruout the earlier work, two other commercial brands of gelatin have been used: one prepared for bacteriological work by the U. S. Glue Co. of Milwaukee; the other, called “ Bactogelatin,” prepared by the Digestive Ferments Co. of Detroit. Both

« PreviousContinue »