January 16, 1914] 



SCIENCE 



81 



the same." The comparative fixity of the 

 more strictly pathogenic bacteria is a strik- 

 ing illustration of this tendency. 



In many instances we find that indi- 

 vidual strains of bacteria exhibit an extra- 

 ordinary uniformity in minor character- 

 istics even when (or perhaps particularly 

 when) cultivated for long periods on arti- 

 ficial media in the laboratory. The two 

 paratyphoid strains, A and B, described 

 above offer a striking instance of this. The 

 mean acid production of the two strains 

 was respectively 1.4 and 1.6 per cent, nor- 

 mal, differing only by .2 per cent, normal 

 and the fluctuating variations, extending 

 over a range of over 1.0 per cent, in each 

 case, far exceed the mean difference be- 

 tween the strains. Yet subcultures show 

 each strain, as a strain, breeding true to 

 its characteristic. We find slight differ- 

 ences in resistance to unfavorable physical 

 condition or to the action of some chemical 

 disinfectant transmitted unchanged in a 

 particular strain for generation after gen- 

 eration. 



As a matter of fact indeed it is not alter- 

 ations in the characters of bacteria while 

 we are studying them which generally trou- 

 ble us, but the fact that as we isolate these 

 organisms in nature we find that antece- 

 dent variations have produced a bewil- 

 dering confusion of slightly differing varie- 

 ties or races or strains. Between well- 

 marked types like B. coli and B. alcaligenes 

 is a series of forms, each one differing but 

 slightly from its neighbor, but together al- 

 most completely bridging the gap between 

 the two extremes. The more refined our 

 methods of bio-chemical examination, the 

 more the types are multiplied, and the 

 more hopeless is the confusion. When Gor- 

 don (1905) applied his nine tests to 300 

 different strains of streptococci, he found 

 48 different combinations of reactions, and 

 MaeConkey (1909) records 36 different 

 varieties of colon bacilli characterized by 



particular combinations of his seven tests 

 To call each distinguishable strain having 

 definite bio-chemical properties a species 

 and to give it a name of its own, is quite 

 out of the question. To ignore all minor 

 differences and maintain as specific such 

 complex groups as are included under the 

 term B. coli or Str. pyogenes is misleading 

 <ind an effective bar to future progress. 



The first principle, which has proved of 

 prime assistance in the characterization of 

 bacterial types, and which offers a rational 

 compromise between either false unity or 

 bewildering multiplicity, is the recognition 

 of the fact that types which occur com- 

 monly among bacteria as they are found in 

 nature are of greater systematic importance 

 than those which occur rarely and occa- 

 sionally. Of course from one standpoint 

 every inheritable protoplasmic variant 

 which exists is of equal importance with 

 every other. For practical purposes, how- 

 ever, we must recognize certain types as 

 " species " or " varieties " even though 

 they may sometimes intergrade. Among 

 the higher plants and animals such syste- 

 matic units are usually recognized on the 

 basis of discontinuity in some definite 

 character. The more refined methods of 

 biometry have however revealed another 

 grade of kinds (I use this word to avoid 

 the artificial implications of " species " or 

 "variety"), marked by relative rather 

 than absolute discontinuity. Frequently 

 the measurement of some particular differ- 

 ential character and the plotting of a " fre- 

 quency polygon," with grades of the 

 chosen character as abscissae and the pro- 

 portion of individuals showing each grade 

 as ordinates, shows a curve with two dis- 

 tinct peaks, a bimodal curve. The studies 

 reported by Bateson on the length of the 

 cephalic horns of' the rhinoceros beetle 

 and on the forceps length of the earwig 

 and De Vries's observations on the petals of 

 a chrysanthemum, are excellent examples. 



